INDEX

Explanations

mentions of the name "Henry" along with associated historical figures

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

ramid

-0.99

awaru

-0.95

ntil

-0.94

ãĥ£

-0.90

*/(

-0.90

cheon

-0.85

umbai

-0.81

olulu

-0.81

alion

-0.79

akening

-0.79

POSITIVE LOGITS

 Rollins

1.11

 Kissinger

1.11

 VIII

1.07

 Ford

1.01

Tud

0.98

Wad

0.92

 Miller

0.88

shire

0.85

 Wink

0.85

Ell

0.81

Activations Density 0.019%

mentions of the name "Henry" along with associated historical figures

No Comments

No Known Activations