INDEX

Explanations

phrases indicating optimal times for action or consideration

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

ancies

-0.74

arser

-0.73

iries

-0.70

uded

-0.68

urat

-0.66

usters

-0.64

leys

-0.63

uthor

-0.62

qus

-0.61

teness

-0.61

POSITIVE LOGITS

 opportunity

0.98

opp

0.83

 ripe

0.73

 introducing

0.72

FORE

0.71

for

0.69

 attRot

0.68

 sleeper

0.66

NING

0.66

fro

0.66

Activations Density 0.100%

phrases indicating optimal times for action or consideration

No Comments

No Known Activations