INDEX

Explanations

technical terms and specialized jargon related to various fields of study

oai_token-act-pair · gpt-3.5-turbo

terms related to scientific concepts and classifications

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olulu

-0.68

ebus

-0.63

 faculties

-0.60

 subcommittee

-0.59

 confir

-0.58

 coerc

-0.57

ividual

-0.57

 gore

-0.57

iston

-0.56

undai

-0.56

POSITIVE LOGITS

 Bastard

0.80

"—

0.79

LOCK

0.68

Class

0.66

lite

0.65

0.64

Dad

0.62

iah

0.61

MAP

0.60

"?

0.60

Activations Density 0.512%

technical terms and specialized jargon related to various fields of study

terms related to scientific concepts and classifications

No Comments

No Known Activations