INDEX

Explanations

references to violent actions or incidents

oai_token-act-pair · gpt-3.5-turbo

occurrences of the verb "was" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

inav

-0.79

 ç¥ŀ

-0.76

poll

-0.75

 Critics

-0.72

 entails

-0.69

 Which

-0.69

ocracy

-0.68

olon

-0.66

 Capitalism

-0.66

 univers

-0.66

POSITIVE LOGITS

 arra

0.97

 hospitalized

0.96

 briefed

0.91

 taken

0.89

 spotted

0.89

 flown

0.87

 arrested

0.86

 alerted

0.86

 ordered

0.85

 supposed

0.85

Activations Density 0.388%

references to violent actions or incidents

occurrences of the verb "was" in various contexts

No Comments

No Known Activations