INDEX

Explanations

phrases indicating comparisons or examples

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istical

-0.83

idate

-0.78

fixed

-0.77

olate

-0.76

gap

-0.76

ettlement

-0.76

nown

-0.75

iminary

-0.75

enser

-0.72

ensibly

-0.72

POSITIVE LOGITS

 Louie

0.92

 Franz

0.89

 Forrest

0.88

 Alfred

0.86

 Jasper

0.86

 Sergio

0.86

 Clive

0.86

 Clint

0.85

 Cowboy

0.85

 Leonardo

0.84

Activations Density 0.071%

phrases indicating comparisons or examples

No Comments

No Known Activations

phrases indicating comparisons or examples

No Comments

No Known Activations