INDEX

Explanations

programming-related structures and functions in code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 disk

-0.17

ollen

-0.15

rum

-0.15

Ø®Ø´

-0.15

ÙĪÙĪ

-0.15

 Schn

-0.14

 Abrams

-0.14

 Nichols

-0.14

urs

-0.14

 jean

-0.14

POSITIVE LOGITS

isel

0.17

elts

0.16

eper

0.16

aga

0.15

 collection

0.15

 Ð²Ð´

0.15

agus

0.15

adla

0.15

agas

0.15

adel

0.14

Activations Density 0.165%

programming-related structures and functions in code

No Comments

No Known Activations