INDEX

Explanations

programming-related terms and structures in code snippets

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

terdam

-0.16

erdem

-0.14

eum

-0.14

 stra

-0.14

ersonic

-0.14

obile

-0.14

ilim

-0.14

ÑģÐ¾Ð»

-0.14

illion

-0.13

vik

-0.13

POSITIVE LOGITS

0.15

roker

0.15

ÑĢÐ¾Ð¶

0.14

("~/

0.14

 Duffy

0.14

lec

0.14

 foreground

0.13

yun

0.13

DL

0.13

anko

0.13

Activations Density 0.071%

programming-related terms and structures in code snippets

No Comments

No Known Activations