INDEX

Explanations

words related to early activities or stages in various contexts

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

ittee

-0.77

retched

-0.74

rawled

-0.73

atism

-0.69

SourceFile

-0.68

md

-0.68

ICLE

-0.66

ingen

-0.65

edIn

-0.65

laws

-0.64

POSITIVE LOGITS

 adop

1.05

 stages

1.01

 indications

0.96

 morning

0.93

 drafts

0.89

 mornings

0.88

 onset

0.85

 phases

0.83

 childhood

0.80

 twenties

0.79

Activations Density 0.346%

words related to early activities or stages in various contexts

No Comments

No Known Activations