INDEX

Explanations

multiple instances of the word "return."

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

ussen

-0.86

 Parenthood

-0.81

Cola

-0.71

esson

-0.69

vern

-0.66

aughtered

-0.66

creen

-0.66

essen

-0.64

 disse

-0.63

rotein

-0.61

POSITIVE LOGITS

ees

1.17

home

0.80

 home

0.76

able

0.76

 postage

0.73

ee

0.71

porting

0.69

prise

0.69

ance

0.68

ported

0.68

Activations Density 0.017%

multiple instances of the word "return."

No Comments

No Known Activations