INDEX

Explanations

the word "case" with a high level of activation

oai_token-act-pair · gpt-3.5-turbo

repetitive phrases indicating a sense of assertion or claim

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

urses

-0.78

 todd

-0.77

inders

-0.76

ĪĴ

-0.76

IRE

-0.72

ursday

-0.72

arding

-0.72

ard

-0.72

ardo

-0.70

dropping

-0.68

POSITIVE LOGITS

 anymore

0.74

ioned

0.72

 scenario

0.71

pheus

0.71

 whatsoever

0.71

liest

0.69

,,,,

0.69

 Isles

0.67

lla

0.67

pring

0.65

Activations Density 0.025%

the word "case" with a high level of activation

repetitive phrases indicating a sense of assertion or claim

No Comments

No Known Activations