INDEX

Explanations

symbols or formatting elements used for separation or emphasis in text

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

Peb

-0.76

perture

-0.72

 graft

-0.68

 scope

-0.67

ttes

-0.65

 redes

-0.64

tti

-0.63

wd

-0.63

olor

-0.62

 strap

-0.62

POSITIVE LOGITS

——

1.27

————

1.17

—-

1.04

ĸļ

1.04

————————————————

1.03

————————

1.03

---------

0.90

ł

0.87

ãĤ¦ãĤ¹

0.84

-+

0.82

Activations Density 0.004%

symbols or formatting elements used for separation or emphasis in text

No Comments

No Known Activations