INDEX

Explanations

references to programming and technical terms 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 @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

vaÅ¾

-0.14

luÅ¾

-0.14

 åĳ

-0.14

 Morrison

-0.14

 Goldberg

-0.13

ãģ¡ãĤī

-0.13

Æ¯

-0.13

Enumeration

-0.13

 accommod

-0.13

viron

-0.13

POSITIVE LOGITS

tes

0.15

AGES

0.15

æŃ²

0.15

.inject

0.15

 Vita

0.15

cade

0.15

TES

0.14

HEST

0.14

aidu

0.14

Activations Density 0.002%

references to programming and technical terms in code snippets

No Comments

No Known Activations