INDEX

Explanations

phrases related to denials or exclusions

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

srf

-0.72

tein

-0.68

 Presence

-0.56

 ancest

-0.55

 assorted

-0.54

 Lanka

-0.52

è¦ļéĨĴ

-0.51

odium

-0.50

 smack

-0.50

 widening

-0.50

POSITIVE LOGITS

 whatsoever

1.04

 necessarily

0.98

 ever

0.91

theless

0.90

 EVER

0.89

conom

0.88

 bothered

0.85

 anymore

0.85

 dime

0.85

ĸļ

0.83

Activations Density 0.568%

phrases related to denials or exclusions

No Comments

No Known Activations