INDEX

Explanations

sections of code spanning multiple lines that perform operations related to data manipulation or flow control

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

imson

-0.15

.Library

-0.15

ICODE

-0.15

ionales

-0.15

pill

-0.14

iges

-0.14

ì»¤

-0.14

zell

-0.14

 proof

-0.14

Cur

-0.14

POSITIVE LOGITS

çĵľ

0.15

amba

0.14

Les

0.14

jit

0.14

abble

0.14

 Olsen

0.14

rown

0.14

azing

0.14

0.13

astery

0.13

Activations Density 0.068%

sections of code spanning multiple lines that perform operations related to data manipulation or flow control

No Comments

No Known Activations