INDEX

Explanations

words related to threats, warnings, and concerns

oai_token-act-pair · gpt-3.5-turbo

terms associated with political and security concerns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

oother

-0.79

 Merge

-0.78

 Transfer

-0.78

Plex

-0.77

 awarding

-0.76

 graft

-0.76

glomer

-0.75

assisted

-0.74

fficiency

-0.73

transfer

-0.72

POSITIVE LOGITS

 fears

1.69

 warnings

1.62

 alarm

1.59

 fear

1.59

 threat

1.56

 warning

1.55

 panic

1.54

 threats

1.54

 worry

1.51

 scares

1.51

Activations Density 0.702%

words related to threats, warnings, and concerns

terms associated with political and security concerns

No Comments

No Known Activations