INDEX

Explanations

phrases related to assessing and evaluating different choices or courses of action

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

oned

-0.67

Die

-0.66

cop

-0.65

wig

-0.63

bug

-0.63

oning

-0.63

 sweat

-0.62

Writ

-0.61

ograph

-0.61

tein

-0.60

POSITIVE LOGITS

 options

1.17

ensical

1.07

 available

1.03

etting

1.00

cale

0.98

 abound

0.92

 choices

0.92

pring

0.88

 Options

0.88

perty

0.84

Activations Density 0.062%

phrases related to assessing and evaluating different choices or courses of action

No Comments

No Known Activations