INDEX

Explanations

declarations of official actions or decisions

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

heit

-0.74

boro

-0.68

 favorably

-0.66

 utilizing

-0.65

ĪĴ

-0.63

reetings

-0.61

kees

-0.60

 hydra

-0.59

 Versus

-0.59

hemoth

-0.58

POSITIVE LOGITS

Scroll

0.96

Labour

0.89

Shape

0.77

However

0.77

 Inqu

0.74

poll

0.73

Asked

0.71

Speaking

0.70

Writing

0.70

 Labour

0.68

Activations Density 0.219%

declarations of official actions or decisions

No Comments

No Known Activations