INDEX

Explanations

phrases describing or outlining plans, cases, issues, rules, or events

oai_token-act-pair · gpt-3.5-turbo

statements that outline or describe plans, issues, or situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

wen

-0.72

tackle

-0.66

ickle

-0.65

wei

-0.65

frog

-0.65

erity

-0.63

WT

-0.63

 Uriel

-0.62

 Adults

-0.61

oso

-0.60

POSITIVE LOGITS

 rationale

1.19

 reasoning

1.09

 criteria

1.06

 reasons

1.02

 particulars

1.02

 procedures

1.01

 virtues

1.01

 workings

1.00

 procedure

1.00

 specifics

0.99

Activations Density 0.351%

phrases describing or outlining plans, cases, issues, rules, or events

statements that outline or describe plans, issues, or situations

No Comments

No Known Activations