INDEX

Explanations

instances of the word "abe."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

phis

-1.07

neapolis

-0.80

ivities

-0.79

iosity

-0.76

ophical

-0.74

 Beir

-0.73

ivity

-0.73

speak

-0.72

stream

-0.71

prus

-0.71

POSITIVE LOGITS

zz

1.01

legates

0.86

legate

0.82

Ã±

0.81

Ã§a

0.81

cki

0.80

cca

0.77

Ca

0.77

FORE

0.77

0.76

Activations Density 0.010%

instances of the word "abe."

No Comments

No Known Activations