INDEX

Explanations

occurrences of specific characters or punctuation marks, particularly apostrophes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

 wallets

-0.68

 Wash

-0.68

 peanuts

-0.65

othy

-0.60

 Morse

-0.60

 stacks

-0.59

 Livingston

-0.59

 Lars

-0.59

 Avery

-0.58

 Crosby

-0.58

POSITIVE LOGITS

oeuv

1.02

Ã©t

0.97

eros

0.89

euro

0.88

avez

0.88

ava

0.79

Ãª

0.79

hist

0.78

esp

0.78

hab

0.77

Activations Density 0.006%

occurrences of specific characters or punctuation marks, particularly apostrophes

No Comments

No Known Activations