INDEX

Explanations

coding-related syntax and structure

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

/apt

-0.15

 Needed

-0.15

 endemic

-0.14

endale

-0.14

 needed

-0.14

 Gand

-0.14

ponse

-0.14

issement

-0.13

STRU

-0.13

ãĢħ

-0.13

POSITIVE LOGITS

 require

0.45

require

0.40

(require

0.32

	require

0.31

 requiring

0.29

 requires

0.29

.require

0.29

 requ

0.27

 Require

0.26

=require

0.25

Activations Density 0.043%

coding-related syntax and structure

No Comments

No Known Activations