INDEX

Explanations

words related to accidents, injuries, and safety tips

oai_token-act-pair · gpt-3.5-turbo

terms related to various social issues, events, and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

$.

-0.66

rather

-0.60

}.

-0.58

etc

-0.58

ļéĨĴ

-0.57

cko

-0.55

ĸļ

-0.55

ensis

-0.54

Interstitial

-0.53

instead

-0.53

POSITIVE LOGITS

 consists

0.63

 extends

0.58

 refers

0.58

 comprises

0.53

has

0.52

 adjusts

0.51

 survives

0.50

 belongs

0.49

 consisted

0.49

 involves

0.49

Activations Density 1.396%

words related to accidents, injuries, and safety tips

terms related to various social issues, events, and conditions

No Comments

No Known Activations