INDEX

Explanations

scientific terms and specific medical conditions or medications

oai_token-act-pair · gpt-3.5-turbo

letters or abbreviations related to specific titles and entities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ongyang

-0.74

 brim

-0.66

osate

-0.66

ccording

-0.64

MF

-0.61

ecause

-0.60

halla

-0.59

icians

-0.59

icipated

-0.59

Chem

-0.58

POSITIVE LOGITS

enment

0.99

ciating

0.74

ĸļ

0.73

 Choi

0.71

isen

0.67

igans

0.66

west

0.65

wikipedia

0.64

hyde

0.63

Ruk

0.62

Activations Density 0.301%

scientific terms and specific medical conditions or medications

letters or abbreviations related to specific titles and entities

No Comments

No Known Activations