INDEX

Explanations

phrases related to climactic moments or revelations in a story

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

ICLE

-0.72

holder

-0.69

 Saud

-0.66

ILD

-0.64

REDACTED

-0.60

fold

-0.60

 Blessed

-0.59

manship

-0.58

 Gutenberg

-0.58

wright

-0.58

POSITIVE LOGITS

pse

1.21

bers

1.14

entary

1.11

ber

1.09

bered

1.03

pson

1.01

actic

0.99

byss

0.97

onial

0.96

psy

0.95

Activations Density 0.028%

phrases related to climactic moments or revelations in a story

No Comments

No Known Activations