INDEX

Explanations

Sequences of related terms, often categories or options, separated by slashes. These frequently represent choices, versions, or specifications within a system or context.

eleuther_acts_top20 · gemini-1.5-flash Triggered by @aserino3

specific technical terms and numbers

oai_token-act-pair · gpt-3.5-turbo

terms related to classifications, sizes, and various forms of media or content

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ographers

-0.64

 braking

-0.64

 scrut

-0.63

 Woods

-0.61

 sworn

-0.61

 Prescott

-0.60

displayText

-0.60

urtles

-0.56

DonaldTrump

-0.55

matically

-0.55

POSITIVE LOGITS

/,

1.19

 combo

1.08

 Ratio

1.05

 ratio

1.05

/)

1.00

/?

0.97

/-

0.96

0.93

/+

0.91

/"

0.90

Activations Density 0.144%

Sequences of related terms, often categories or options, separated by slashes. These frequently represent choices, versions, or specifications within a system or context.

specific technical terms and numbers

terms related to classifications, sizes, and various forms of media or content

No Comments

No Known Activations