INDEX

Explanations

terms related to giving permission or approval, such as "clearance."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "clearance"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

amed

-0.71

eva

-0.69

sn

-0.66

repl

-0.66

ram

-0.66

Sn

-0.66

rame

-0.65

LI

-0.65

hed

-0.65

sg

-0.63

POSITIVE LOGITS

 clearance

1.69

 waivers

0.93

 clears

0.84

 cleared

0.83

ACP

0.77

 certs

0.76

 paperwork

0.76

Ĥª

0.75

 approvals

0.73

 clearing

0.73

Activations Density 0.008%

terms related to giving permission or approval, such as "clearance."

occurrences of the word "clearance"

No Comments

No Known Activations