INDEX

Explanations

phrases related to taking action or control

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

ibel

-0.78

icol

-0.73

lain

-0.71

voy

-0.71

esa

-0.71

mint

-0.71

merce

-0.69

heid

-0.69

SPONSORED

-0.68

cius

-0.67

POSITIVE LOGITS

 levers

1.02

 aggro

0.93

awa

0.89

 punches

0.88

 away

0.86

awed

0.83

 strings

0.83

 weeds

0.82

 hamstring

0.81

pull

0.80

Activations Density 0.500%

phrases related to taking action or control

No Comments

No Known Activations