INDEX

Explanations

frequent references to recurring themes or behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Unchecked

-0.17

sert

-0.16

ixo

-0.16

acades

-0.15

factory

-0.15

unas

-0.15

WSC

-0.14

owanie

-0.14

/******/

-0.14

lers

-0.14

POSITIVE LOGITS

-times

0.23

entimes

0.21

-used

0.18

 times

0.16

 xuyÃªn

0.15

heimer

0.15

obre

0.14

eda

0.14

IGHL

0.14

ìĶ©

0.14

Activations Density 0.037%

frequent references to recurring themes or behaviors

No Comments

No Known Activations