INDEX

Explanations

similes using the word "like"

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

Published

-0.89

inion

-0.77

hiba

-0.74

idates

-0.72

inas

-0.71

ittal

-0.71

endiary

-0.69

iets

-0.68

ione

-0.67

overy

-0.67

POSITIVE LOGITS

lihood

1.87

lier

1.34

liest

1.33

liness

0.98

 minded

0.97

minded

0.94

 ours

0.93

 wildfire

0.85

 clock

0.79

 yours

0.79

Activations Density 2.327%

similes using the word "like"

No Comments

No Known Activations