INDEX

Explanations

words related to people in specific roles or positions

oai_token-act-pair · gpt-3.5-turbo

words related to students and participants 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

cer

-0.71

 Kuala

-0.68

rawdownloadcloneembedreportprint

-0.65

OOK

-0.64

 Variety

-0.62

tein

-0.62

Thom

-0.61

pour

-0.57

south

-0.57

Kov

-0.57

POSITIVE LOGITS

 selves

0.98

 counterparts

0.94

folk

0.85

issance

0.85

mates

0.81

 brethren

0.80

hip

0.79

ervative

0.78

 holdings

0.69

iphate

0.68

Activations Density 0.151%

words related to people in specific roles or positions

words related to students and participants in various contexts

No Comments

No Known Activations