INDEX

Explanations

mentions of fountains, potentially with different levels of emphasis

oai_token-act-pair · gpt-3.5-turbo

references to fountains or related terms in various contexts

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

 Sisters

-0.65

 Ibid

-0.64

Alien

-0.63

oan

-0.62

 bang

-0.62

LU

-0.61

 Mald

-0.60

 Dickens

-0.59

culosis

-0.59

 Hung

-0.59

POSITIVE LOGITS

 fountain

1.07

sonian

1.03

issance

0.97

shire

0.92

 Fountain

0.88

 pens

0.87

ocity

0.84

ĺħ

0.82

atre

0.80

tailed

0.77

Activations Density 0.020%

mentions of fountains, potentially with different levels of emphasis

references to fountains or related terms in various contexts

No Comments

No Known Activations