INDEX

Explanations

phrases indicating quantity or frequency

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

adjectives describing different characteristics or states

oai_token-act-pair · gpt-3.5-turbo

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

alion

-0.64

icism

-0.63

culosis

-0.63

bara

-0.62

CRE

-0.62

erity

-0.59

APTER

-0.58

EMA

-0.57

 Barrett

-0.56

cation

-0.56

POSITIVE LOGITS

 themselves

0.89

 interchangeable

0.79

abouts

0.78

 selves

0.72

 substitutes

0.70

 equivalents

0.70

 extensions

0.69

types

0.69

 outl

0.68

 exceptions

0.66

Activations Density 0.924%

phrases indicating quantity or frequency

adjectives describing different characteristics or states

No Comments

No Known Activations