INDEX

Explanations

punctuation marks and numerical values within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

Ans

-0.15

berman

-0.15

Ð½Ð°Ð´

-0.14

ewn

-0.14

 <!--[

-0.14

hood

-0.14

Ã©l

-0.13

æ¹¾

-0.13

olla

-0.13

itude

-0.13

POSITIVE LOGITS

RBI

0.20

 Assist

0.16

 advancement

0.16

 TEAM

0.16

 advanced

0.15

 Team

0.15

@student

0.14

advanced

0.14

ãĤ¿ãĤ¤

0.14

Activations Density 0.000%

punctuation marks and numerical values within the text

No Comments

No Known Activations