INDEX

Explanations

phrases related to promoting or generating interest in a topic or idea

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

anwhile

-0.75

amazon

-0.69

TPPStreamerBot

-0.67

iw

-0.66

bard

-0.66

sych

-0.65

forge

-0.63

HUD

-0.63

ß

-0.63

cius

-0.62

POSITIVE LOGITS

 concessions

0.95

 victories

0.89

 revenge

0.82

 scraps

0.79

 havoc

0.79

 extra

0.77

 strikeouts

0.75

 laughs

0.75

 bits

0.73

 vengeance

0.73

Activations Density 0.079%

phrases related to promoting or generating interest in a topic or idea

No Comments

No Known Activations

phrases related to promoting or generating interest in a topic or idea

No Comments

No Known Activations