INDEX

Explanations

proper names or nouns related to figures

oai_token-act-pair · gpt-3.5-turbo

words that begin with the letter 'R'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 needle

-0.60

 conveniently

-0.60

 chains

-0.59

ypes

-0.57

lehem

-0.57

 envy

-0.57

 Chains

-0.56

 matched

-0.56

 loophole

-0.56

 luck

-0.56

POSITIVE LOGITS

issance

0.90

kefeller

0.88

zl

0.86

earchers

0.77

eway

0.76

restling

0.76

ighters

0.75

ael

0.74

backer

0.73

heed

0.73

Activations Density 0.098%

proper names or nouns related to figures

words that begin with the letter 'R'

No Comments

No Known Activations