INDEX

Explanations

references to functions and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

you

-0.16

å½¹

-0.14

æ²¡

-0.14

Ø´Ùĩ

-0.14

PEnd

-0.14

 rÃ¡d

-0.13

Kit

-0.13

regnum

-0.13

 arms

-0.13

POSITIVE LOGITS

gren

0.17

.NaN

0.14

iaux

0.14

 somebody

0.14

ãģ¨ãĤĤ

0.14

à¸±à¸Ļà¸Ļ

0.13

 Dane

0.13

utan

0.13

raud

0.13

rote

0.13

Activations Density 0.001%

references to functions and processes

No Comments

No Known Activations