INDEX

Explanations

specific numerical data and patterns in structured information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

kes

-0.16

ãĤªãĥª

-0.16

zer

-0.15

erce

-0.15

nesc

-0.14

asje

-0.14

Ð¾ÑĪ

-0.14

ousel

-0.13

Ã¶yle

-0.13

 sole

-0.13

POSITIVE LOGITS

ilog

0.15

omin

0.14

ÑģÑİ

0.14

Eis

0.14

oli

0.14

 warrant

0.13

itzer

0.13

yp

0.13

 Rail

0.13

 ///<

0.13

Activations Density 0.119%

specific numerical data and patterns in structured information

No Comments

No Known Activations