INDEX

Explanations

names or surnames of individuals in various contexts

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

izoph

-0.73

 unfavorable

-0.65

Compared

-0.64

fml

-0.62

ãĥ¼ãĥ³

-0.61

MAT

-0.60

 allowable

-0.59

 dysfunctional

-0.59

 unsustainable

-0.58

BIL

-0.58

POSITIVE LOGITS

Jr

1.37

III

1.10

Sr

1.09

zzi

1.03

aka

1.01

baum

0.99

tsky

0.95

iewicz

0.93

QC

0.90

ieri

0.90

Activations Density 0.200%

names or surnames of individuals in various contexts

No Comments

No Known Activations