INDEX

Explanations

programming-related keywords and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

isseur

-0.16

gesi

-0.16

abay

-0.16

-0.15

rac

-0.15

lash

-0.15

.TryParse

-0.14

Emm

-0.14

olet

-0.14

omial

-0.14

POSITIVE LOGITS

ãĥ¼ãĥį

0.16

orsi

0.15

ëĵ¯

0.14

ä»ķ

0.14

ati

0.14

 DÃ¢n

0.14

èĬĤ

0.14

 widely

0.13

avor

0.13

 facial

0.13

Activations Density 0.064%

programming-related keywords and structures

No Comments

No Known Activations