INDEX

Explanations

references to car brands and models

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to luxury brands and specific 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

ingly

-0.80

inia

-0.73

//[

-0.71

 GOODMAN

-0.70

listed

-0.70

é¾įå

-0.70

rider

-0.67

########

-0.66

avis

-0.66

VID

-0.65

POSITIVE LOGITS

oshenko

0.87

Ã¶n

0.87

roleum

0.81

asus

0.80

ongyang

0.79

insula

0.77

ainted

0.77

pload

0.76

akura

0.75

neum

0.75

Activations Density 0.031%

references to car brands and models

proper nouns related to luxury brands and specific individuals

No Comments

No Known Activations