INDEX

Explanations

words related to action or processes, particularly focusing on verbs and their variations

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

FUL

-1.07

Ô

-0.90

é¾įå

-0.75

cipled

-0.74

stellar

-0.73

)=(

-0.72

landish

-0.72

 Continued

-0.70

çĶ°

-0.68

Pod

-0.68

POSITIVE LOGITS

asers

0.97

arers

0.93

ippers

0.91

ancies

0.90

acts

0.89

glers

0.88

ipper

0.88

ricks

0.88

amps

0.88

odes

0.87

Activations Density 0.225%

words related to action or processes, particularly focusing on verbs and their variations

No Comments

No Known Activations