INDEX

Explanations

verbs that imply addressing or dealing with a specific topic or issue

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "addressed."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ischer

-0.76

Fav

-0.71

 Bret

-0.71

bp

-0.66

axis

-0.65

batch

-0.64

cop

-0.63

Du

-0.63

thing

-0.63

aren

-0.63

POSITIVE LOGITS

 addressed

0.95

xual

0.92

 addressing

0.92

 address

0.78

 addresses

0.77

address

0.74

tnc

0.72

 Address

0.71

ãĤĮ

0.69

Address

0.68

Activations Density 0.012%

verbs that imply addressing or dealing with a specific topic or issue

occurrences of the word "addressed."

No Comments

No Known Activations