INDEX

Explanations

phrases involving proportions or comparisons

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

body

-0.60

Task

-0.55

wx

-0.55

machine

-0.54

paren

-0.54

 destro

-0.53

irez

-0.52

kamp

-0.52

assembly

-0.52

tera

-0.51

POSITIVE LOGITS

 course

0.69

course

0.68

 sorts

0.67

Course

0.58

sted

0.57

 whom

0.57

uits

0.56

us

0.55

urnal

0.54

 Hers

0.54

Activations Density 0.053%

phrases involving proportions or comparisons

No Comments

No Known Activations