INDEX

Explanations

references to specific numerical rankings or positions

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

 destro

-0.69

 totality

-0.69

 muted

-0.68

endish

-0.63

 conj

-0.61

lyn

-0.61

 composing

-0.61

 awoken

-0.60

 cruc

-0.59

 suppressed

-0.58

POSITIVE LOGITS

xious

1.02

zzle

0.97

obs

0.87

isy

0.83

emi

0.83

AH

0.83

onday

0.82

Such

0.79

ises

0.78

DIV

0.78

Activations Density 0.016%

references to specific numerical rankings or positions

No Comments

No Known Activations