INDEX

Explanations

words related to presenting and discussing arguments

oai_token-act-pair · gpt-3.5-turbo

mentions of arguments or discussions on various topics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eco

-0.75

cler

-0.68

 livest

-0.67

 Atomic

-0.66

aches

-0.65

oho

-0.65

ookie

-0.64

isner

-0.63

 Seym

-0.62

ummer

-0.61

POSITIVE LOGITS

uments

1.16

ative

1.09

ument

0.94

 arguments

0.93

 argument

0.88

 against

0.86

 arguing

0.82

eering

0.80

acle

0.77

ļéĨĴ

0.76

Activations Density 0.020%

words related to presenting and discussing arguments

mentions of arguments or discussions on various topics

No Comments

No Known Activations