INDEX

Explanations

references to specific types of activities or events

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "sessions."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

zo

-0.72

Uz

-0.64

ARDS

-0.62

FORE

-0.61

ACTED

-0.60

 generous

-0.60

tymology

-0.58

Constructed

-0.57

OV

-0.57

dup

-0.57

POSITIVE LOGITS

 sessions

1.38

 session

1.10

ongs

0.94

ional

0.91

hops

0.90

ilver

0.90

è¦ļéĨĴ

0.88

creen

0.88

 Session

0.87

essions

0.84

Activations Density 0.008%

references to specific types of activities or events

occurrences of the word "sessions."

No Comments

No Known Activations