INDEX

Explanations

expressions indicating the passage of time or significant changes in state

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

 Leilan

-0.74

itton

-0.71

arium

-0.62

ugi

-0.61

 assorted

-0.60

 instead

-0.58

ngth

-0.58

esm

-0.56

leigh

-0.54

ortment

-0.54

POSITIVE LOGITS

 anymore

1.30

nor

1.09

 necessarily

0.91

 anywhere

0.87

 slightest

0.87

yet

0.87

any

0.83

 anything

0.79

 anytime

0.70

bole

0.69

Activations Density 0.135%

expressions indicating the passage of time or significant changes in state

No Comments

No Known Activations