INDEX

Explanations

phrases related to discoveries or realizations

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

idium

-0.73

 concess

-0.64

puff

-0.63

perty

-0.63

panic

-0.62

inion

-0.60

rolet

-0.59

oiler

-0.59

VP

-0.58

brance

-0.58

POSITIVE LOGITS

 fault

0.80

ãĤ¤ãĥĪ

0.79

ered

0.78

ibility

0.78

 loopholes

0.77

ById

0.76

 traces

0.75

 objectionable

0.74

out

0.74

omething

0.72

Activations Density 1.054%

phrases related to discoveries or realizations

No Comments

No Known Activations