INDEX

Explanations

mentions of political or governmental policies

oai_token-act-pair · gpt-3.5-turbo

references to government policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

issan

-0.84

ITNESS

-0.82

lighting

-0.76

Ort

-0.74

 Sabha

-0.73

sembly

-0.73

Rocket

-0.73

CLASSIFIED

-0.72

athan

-0.71

Sud

-0.70

POSITIVE LOGITS

 policies

1.32

 policy

1.06

 Policies

1.05

 prescriptions

0.99

 stances

0.90

policy

0.84

 stance

0.83

 preferences

0.82

olicy

0.81

 directives

0.81

Activations Density 0.014%

mentions of political or governmental policies

references to government policies

No Comments

No Known Activations