INDEX

Explanations

programming-related keywords and constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

AllowAnonymous

-0.16

utton

-0.15

ãĥ

-0.15

ÑħÐ¾Ð²

-0.15

otos

-0.14

oton

-0.14

ingu

-0.14

ahat

-0.14

.pair

-0.14

innacle

-0.14

POSITIVE LOGITS

rium

0.14

ITLE

0.14

ilyn

0.14

ãĥ¼ãĥ«

0.14

eteor

0.14

ahun

0.14

lt

0.13

pline

0.13

 Hill

0.13

ëĿ½

0.13

Activations Density 0.002%

programming-related keywords and constructs

No Comments

No Known Activations