INDEX

Explanations

adjectives ending in 'able' or 'ible'

oai_token-act-pair · gpt-3.5-turbo

specific verbs and nouns that indicate actions or processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

tsky

-0.61

lang

-0.61

odan

-0.61

etter

-0.60

EStream

-0.58

CHO

-0.58

DragonMagazine

-0.57

NRS

-0.57

 Anon

-0.57

creen

-0.57

POSITIVE LOGITS

iencies

0.97

ated

0.89

ibly

0.87

ably

0.86

cation

0.85

acy

0.84

icative

0.83

å£«

0.80

ATED

0.80

ately

0.79

Activations Density 0.197%

adjectives ending in 'able' or 'ible'

specific verbs and nouns that indicate actions or processes

No Comments

No Known Activations