INDEX

Explanations

names or terms related to a specific video game or characters within it

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Dra" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.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

acion

-0.73

ulators

-0.71

ARC

-0.71

 GOODMAN

-0.70

ALLY

-0.68

LESS

-0.64

amide

-0.63

 Bermuda

-0.63

 surrog

-0.62

 Contra

-0.61

POSITIVE LOGITS

ven

1.08

ught

1.06

fters

1.01

fter

0.95

uth

0.88

cffff

0.88

rys

0.88

han

0.86

ille

0.84

enei

0.84

Activations Density 0.006%

names or terms related to a specific video game or characters within it

mentions of the name "Dra" in various contexts

No Comments

No Known Activations