INDEX

Explanations

instances of the word "ip" in various contexts

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

DEV

-0.72

 EntityItem

-0.72

BIP

-0.70

 Baldwin

-0.66

 dots

-0.65

 Parenthood

-0.63

 Gael

-0.63

abouts

-0.63

 Chatt

-0.61

ASD

-0.60

POSITIVE LOGITS

otent

1.16

artisan

1.05

ede

1.01

osit

0.95

airs

0.92

resent

0.92

uls

0.92

ulse

0.91

ressive

0.88

rocal

0.87

Activations Density 0.009%

instances of the word "ip" in various contexts

No Comments

No Known Activations