INDEX

Explanations

phrases involving the word "for"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

bda

-0.65

querque

-0.64

jab

-0.62

Introduced

-0.61

claimed

-0.61

daq

-0.61

 Originally

-0.60

til

-0.59

à©

-0.59

Against

-0.59

POSITIVE LOGITS

us

1.05

bidden

1.01

 someone

0.97

 them

0.97

me

0.92

geries

0.92

 anyone

0.92

him

0.88

 individuals

0.83

 policymakers

0.82

Activations Density 0.092%

phrases involving the word "for"

No Comments

No Known Activations