INDEX

Explanations

the name "Gene" mentioned in the text

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

hops

-0.78

akening

-0.76

footed

-0.75

yrim

-0.74

ahime

-0.74

ecause

-0.73

ango

-0.69

lain

-0.68

ategory

-0.67

etheless

-0.67

POSITIVE LOGITS

alogy

1.51

alog

1.31

etics

1.11

etically

0.92

Gene

0.87

etic

0.86

ocide

0.82

 Simmons

0.79

elsen

0.79

va

0.78

Activations Density 0.014%

the name "Gene" mentioned in the text

No Comments

No Known Activations