INDEX

Explanations

references to typesetting or document formatting commands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

å¼ĺ

-0.15

ftime

-0.15

ush

-0.14

THR

-0.14

Thr

-0.14

ãĥĥãĤ·ãĥ¥

-0.14

dez

-0.13

ÑģÐ»Ð¸

-0.13

 Hole

-0.13

valuate

-0.13

POSITIVE LOGITS

UTF

0.17

 paper

0.16

lip

0.15

 Margin

0.15

 margin

0.14

EventListener

0.14

-paper

0.14

UME

0.14

Ø³Ø§ÙĨ

0.14

Margin

0.13

Activations Density 0.012%

references to typesetting or document formatting commands

No Comments

No Known Activations