INDEX

Explanations

phrases related to government and political systems

oai_token-act-pair · gpt-3.5-turbo

references to governmental actions and decisions

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

 Powered

-0.63

Fug

-0.59

 Gravity

-0.55

Hok

-0.55

 Highlander

-0.55

 Later

-0.54

 Stoke

-0.54

Ud

-0.53

 onwards

-0.52

 Showdown

-0.52

POSITIVE LOGITS

 acknowled

0.63

Closure

0.61

ateral

0.60

ONSORED

0.57

 understatement

0.55

 adultery

0.54

onia

0.54

coni

0.53

phia

0.53

âĢ¢âĢ¢âĢ¢âĢ¢

0.53

Activations Density 0.754%

phrases related to government and political systems

references to governmental actions and decisions

No Comments

No Known Activations