INDEX

Explanations

comparative phrases involving the word "like."

oai_token-act-pair · gpt-4o-mini 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

inion

-0.78

ilic

-0.77

Published

-0.76

inas

-0.74

hiba

-0.74

elight

-0.72

hani

-0.72

ulty

-0.71

byn

-0.70

iets

-0.70

POSITIVE LOGITS

liest

1.10

lihood

1.09

lier

1.02

 wildfire

0.95

 clock

0.90

 crazy

0.89

liness

0.84

minded

0.82

 minded

0.82

 usual

0.73

Activations Density 0.075%

comparative phrases involving the word "like."

No Comments

No Known Activations