INDEX

Explanations

programming-related commands 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 @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ized

-0.19

hood

-0.18

stown

-0.17

-vous

-0.16

edback

-0.16

boro

-0.16

eenth

-0.16

edReader

-0.16

 latter

-0.16

hip

-0.15

POSITIVE LOGITS

Ùĳ

0.16

uti

0.16

Î½ÏĦ

0.16

 à¹Ĩ

0.15

tek

0.15

ãģ¾ãģŁ

0.15

ÛĮØ³

0.14

na

0.14

usal

0.14

ri

0.14

Activations Density 0.474%

programming-related commands and structures

No Comments

No Known Activations