INDEX

Explanations

references to legislation and policies

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

nat

-0.60

 flanked

-0.59

urai

-0.58

 thanking

-0.57

gently

-0.56

 promptly

-0.56

achine

-0.55

ornia

-0.55

wrong

-0.55

 revenge

-0.53

POSITIVE LOGITS

 counterparts

0.96

 predecessors

0.86

 ones

0.74

 attRot

0.72

().

0.70

nor

0.69

 equivalents

0.69

 predec

0.68

pez

0.66

 preceding

0.65

Activations Density 3.816%

references to legislation and policies

No Comments

No Known Activations