INDEX

Explanations

mentions of specific medical treatments or procedures

oai_token-act-pair · gpt-3.5-turbo

repeated references to the term "pet."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ĸļ

-0.93

ught

-0.63

eve

-0.61

 remed

-0.59

WARE

-0.59

ULTS

-0.58

 fitting

-0.58

govtrack

-0.58

pta

-0.57

WORK

-0.57

POSITIVE LOGITS

lehem

0.92

angelo

0.85

sels

0.84

inas

0.80

iana

0.71

baum

0.70

asus

0.69

isi

0.69

uese

0.68

anski

0.68

Activations Density 0.148%

mentions of specific medical treatments or procedures

repeated references to the term "pet."

No Comments

No Known Activations