INDEX

Explanations

proper nouns referring to people making statements or giving interviews

oai_token-act-pair · gpt-3.5-turbo

reported speech or statements made by individuals

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

VIS

-0.65

amac

-0.64

pee

-0.63

fell

-0.63

ARS

-0.60

ayne

-0.59

isal

-0.58

agra

-0.57

involved

-0.56

arette

-0.56

POSITIVE LOGITS

 thanking

0.68

 aloud

0.67

 plaint

0.67

 bluntly

0.66

 goodbye

0.66

 sarcast

0.64

:]

0.63

aturdays

0.61

 categor

0.61

:"

0.61

Activations Density 0.227%

proper nouns referring to people making statements or giving interviews

reported speech or statements made by individuals

No Comments

No Known Activations