INDEX

Explanations

political and social terms or phrases with strong connotations

oai_token-act-pair · gpt-3.5-turbo

references to political figures 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 @ 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

 signatures

-0.71

reports

-0.70

ernels

-0.67

Hoo

-0.66

sites

-0.66

rooms

-0.65

udos

-0.65

nces

-0.64

 bells

-0.63

facts

-0.63

POSITIVE LOGITS

 standpoint

0.84

 opponent

0.72

 plantation

0.65

 competitor

0.64

 printer

0.64

 colleague

0.64

 sized

0.64

 colony

0.63

 ancestor

0.62

rawn

0.61

Activations Density 0.749%

political and social terms or phrases with strong connotations

references to political figures and social issues

No Comments

No Known Activations