INDEX

Explanations

mentions of winning or victory

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

Factor

-0.66

senal

-0.62

bian

-0.60

uras

-0.60

ossus

-0.57

rouch

-0.57

 periphery

-0.57

erity

-0.56

ria

-0.56

opian

-0.54

POSITIVE LOGITS

now

1.09

nings

1.05

throp

1.00

't

0.98

cest

0.87

ced

0.87

cing

0.81

ipeg

0.80

ests

0.80

hardt

0.79

Activations Density 1.287%

mentions of winning or victory

No Comments

No Known Activations