INDEX

Explanations

feedback and requests for updates related to development issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

Å¡et

-0.15

TeX

-0.14

anc

-0.14

 Lund

-0.14

cence

-0.14

biz

-0.14

 DISP

-0.14

au

-0.14

 dispens

-0.13

"%.

-0.13

POSITIVE LOGITS

Trace

0.17

elo

0.17

eros

0.17

 Trace

0.16

ested

0.15

aghan

0.14

TRACE

0.14

Ø§ÙĨÙĩ

0.14

ocale

0.14

uche

0.13

Activations Density 0.462%

feedback and requests for updates related to development issues

No Comments

No Known Activations