INDEX

Explanations

technical and software-related terms or descriptions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

 Entered

-0.61

 Hait

-0.54

 Murdoch

-0.54

 Nielsen

-0.53

 Hutchinson

-0.53

 Ambro

-0.53

 Ulster

-0.53

 Sears

-0.52

 slog

-0.51

 strapped

-0.51

POSITIVE LOGITS

SourceFile

0.77

iverse

0.77

reciation

0.72

 itself

0.70

brids

0.70

prototype

0.69

stice

0.67

yssey

0.66

iqueness

0.65

ãĥĺãĥ©

0.64

Activations Density 2.491%

technical and software-related terms or descriptions

No Comments

No Known Activations