INDEX

Explanations

phrases related to instructional content or step-by-step procedures

oai_token-act-pair · gpt-3.5-turbo 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

Rab

-0.69

rones

-0.69

 Notting

-0.68

lights

-0.64

erity

-0.63

 Blitz

-0.63

-0.61

 Sons

-0.61

 Desire

-0.61

 Mans

-0.60

POSITIVE LOGITS

 traced

1.09

 viewed

1.08

 easily

1.03

 seen

1.00

 likened

1.00

 construed

0.94

 accessed

0.93

 considered

0.91

 forgiven

0.90

 safely

0.88

Activations Density 0.092%

phrases related to instructional content or step-by-step procedures

No Comments

No Known Activations