INDEX

Explanations

words related to spending time or money

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

named

-0.59

 Cosponsors

-0.59

 Defeat

-0.59

ensen

-0.56

à¼

-0.56

quartered

-0.56

 wont

-0.56

 Parade

-0.55

deal

-0.54

Carter

-0.54

POSITIVE LOGITS

itures

1.04

lav

0.93

 eternity

0.89

IPM

0.87

 winters

0.86

iture

0.84

 money

0.84

 untold

0.82

 hours

0.82

htaking

0.79

Activations Density 0.045%

words related to spending time or money

No Comments

No Known Activations