INDEX

Explanations

organizations or groups with the word "Front" in their name

oai_token-act-pair · gpt-3.5-turbo

references to various political groups or organizations identified by the term "Front."

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

UGH

-0.82

awk

-0.80

STEM

-0.74

tar

-0.74

risome

-0.73

å¥³

-0.73

chell

-0.72

gging

-0.71

cles

-0.71

awks

-0.71

POSITIVE LOGITS

alis

0.88

ugal

0.85

ois

0.81

eous

0.78

icia

0.76

iers

0.76

Front

0.75

ing

0.73

 Against

0.72

ais

0.69

Activations Density 0.017%

organizations or groups with the word "Front" in their name

references to various political groups or organizations identified by the term "Front."

No Comments

No Known Activations