INDEX

Explanations

terms and key phrases that denote specific concepts or definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

tract

-0.15

 <!--[

-0.14

rych

-0.14

 Hampshire

-0.14

Mod

-0.13

holm

-0.13

ep

-0.13

gaard

-0.13

imi

-0.13

ve

-0.13

POSITIVE LOGITS

 McGregor

0.17

èĭĹ

0.16

MBED

0.15

ãĥ¼ãĥĳãĥ¼

0.15

uesta

0.15

itecture

0.15

kre

0.15

ableObject

0.15

lse

0.14

ãĥ¼ãĥĵãĤ¹

0.14

Activations Density 0.030%

terms and key phrases that denote specific concepts or definitions

No Comments

No Known Activations