INDEX

Explanations

indicators of categorization or classification in content

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

åĤ

-0.15

uste

-0.15

bst

-0.14

 Hutchinson

-0.14

edi

-0.14

Ø±Ùģ

-0.14

ovie

-0.14

/umd

-0.13

ä¾Ľ

-0.13

 Reef

-0.13

POSITIVE LOGITS

gro

0.16

 Uncategorized

0.16

aes

0.16

ijn

0.15

pll

0.15

Ã±ana

0.15

 Ãºn

0.15

 krat

0.14

ouv

0.14

ANGO

0.14

Activations Density 0.001%

indicators of categorization or classification in content

No Comments

No Known Activations