INDEX

Explanations

the word "Before" at the beginning of text segments

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

geist

-0.74

aren

-0.72

atic

-0.68

aez

-0.66

SA

-0.64

ingle

-0.64

Offline

-0.63

ãĤ¤

-0.62

Artist

-0.62

among

-0.61

POSITIVE LOGITS

 committing

0.93

 embark

0.90

 sunrise

0.89

 expiration

0.84

 midnight

0.83

 realizing

0.81

 proceeding

0.81

 heading

0.80

 noon

0.79

 sunset

0.77

Activations Density 1.884%

the word "Before" at the beginning of text segments

No Comments

No Known Activations