INDEX

Explanations

statements where agreement or alignment is expressed

oai_token-act-pair · gpt-3.5-turbo

expressions of agreement

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

sembly

-0.73

asus

-0.71

hyde

-0.68

gallery

-0.67

ccording

-0.66

ò

-0.64

oufl

-0.62

inary

-0.62

agin

-0.61

crow

-0.60

POSITIVE LOGITS

 with

0.90

 vehemently

0.85

 unanimously

0.76

 strongly

0.76

SOURCE

0.72

atively

0.70

wi

0.68

 WITH

0.66

 passionately

0.66

ohn

0.65

Activations Density 0.036%

statements where agreement or alignment is expressed

expressions of agreement

No Comments

No Known Activations