INDEX

Explanations

the word "which" 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

LET

-0.76

let

-0.72

politics

-0.69

ahime

-0.66

CLOSE

-0.65

nor

-0.64

grim

-0.63

va

-0.63

quote

-0.63

ct

-0.62

POSITIVE LOGITS

 originated

1.00

 involve

0.91

 contain

0.89

are

0.88

 contribute

0.86

 originate

0.83

 contributed

0.82

 were

0.82

 resulted

0.81

 seem

0.79

Activations Density 0.012%

the word "which" in various contexts

No Comments

No Known Activations