INDEX

Explanations

occurrences of the word "in."

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

NOW

-0.77

CLASSIFIED

-0.76

issance

-0.75

$$

-0.72

gans

-0.71

leeve

-0.71

emis

-0.71

hiba

-0.71

aida

-0.70

cyclopedia

-0.70

POSITIVE LOGITS

 succession

1.05

 conjunction

0.97

 spite

0.96

 front

0.95

 lieu

0.92

 unison

0.92

 finals

0.88

 favor

0.87

 favour

0.84

ked

0.83

Activations Density 0.081%

occurrences of the word "in."

No Comments

No Known Activations