INDEX

Explanations

word sequences related to government actions and decisions

oai_token-act-pair · gpt-3.5-turbo

elements related to legal and governmental processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 destro

-0.59

cla

-0.57

 blades

-0.55

 blond

-0.55

 fascination

-0.55

 Moroc

-0.55

 contrace

-0.54

 fascinated

-0.54

 melted

-0.54

 experien

-0.54

POSITIVE LOGITS

 Legislative

0.82

 appoint

0.81

 appointing

0.80

liament

0.79

 Fiscal

0.79

 Elections

0.79

 Legislation

0.78

displayText

0.78

Senate

0.78

Council

0.77

Activations Density 1.137%

word sequences related to government actions and decisions

elements related to legal and governmental processes

No Comments

No Known Activations