INDEX

Explanations

phrases indicating purpose or intent

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Needs

-0.80

 Appears

-0.72

tis

-0.72

Ake

-0.69

 Clover

-0.68

 Knot

-0.67

 Mead

-0.66

ãĤ¦ãĤ¹

-0.66

Klu

-0.65

Ves

-0.64

POSITIVE LOGITS

 communicate

1.20

 circumvent

1.16

 educate

1.11

 promote

1.10

 enhance

1.07

 introduce

1.05

 organize

1.05

 empower

1.02

 alleviate

1.01

 differentiate

1.01

Activations Density 0.051%

phrases indicating purpose or intent

No Comments

No Known Activations