INDEX

Explanations

phrases related to transformations or changes

oai_token-act-pair · gpt-3.5-turbo

phrases indicating transformation or change

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

orno

-0.67

entimes

-0.64

rolley

-0.64

 indeed

-0.63

 reluct

-0.63

 slowdown

-0.61

 evacuated

-0.60

 timing

-0.59

aque

-0.59

 pressures

-0.58

POSITIVE LOGITS

 adulthood

0.74

ãĤ©

0.71

pload

0.69

âĸĳâĸĳ

0.69

izoph

0.68

clusion

0.67

displayText

0.66

clus

0.65

Role

0.65

obin

0.65

Activations Density 0.064%

phrases related to transformations or changes

phrases indicating transformation or change

No Comments

No Known Activations