INDEX

Explanations

phrases related to function returns in programming

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

-0.79

iece

-0.73

anto

-0.72

enture

-0.72

pai

-0.70

stem

-0.70

keley

-0.69

egal

-0.66

ortium

-0.66

 sabotage

-0.65

POSITIVE LOGITS

 TRUE

0.91

nil

0.80

izable

0.74

 results

0.72

 FALSE

0.71

 value

0.70

 NULL

0.70

VALUE

0.70

ported

0.70

value

0.70

Activations Density 0.017%

phrases related to function returns in programming

No Comments

No Known Activations