INDEX

Explanations

mentions of and sentiments towards anti-government or politically opposing activities

oai_token-act-pair · gpt-3.5-turbo

terms related to political opposition and social issues

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

iosis

-0.81

 srfAttach

-0.80

IENCE

-0.75

LESS

-0.71

OST

-0.68

 invested

-0.68

WIND

-0.65

 FANTASY

-0.63

ahime

-0.62

OIL

-0.61

POSITIVE LOGITS

usterity

0.86

xiety

0.85

 crus

0.78

 Measures

0.76

heid

0.76

effects

0.73

violence

0.71

 sentiment

0.69

byter

0.68

 Activities

0.68

Activations Density 0.066%

mentions of and sentiments towards anti-government or politically opposing activities

terms related to political opposition and social issues

No Comments

No Known Activations