INDEX

Explanations

code snippets and programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

anguard

-0.17

BufferData

-0.16

.Transactional

-0.15

 Broad

-0.15

Ã©sz

-0.14

idata

-0.14

âĿ

-0.14

.xr

-0.13

onte

-0.13

stdafx

-0.13

POSITIVE LOGITS

 Howell

0.16

 Ø§ÙĦØ¨Ø¯

0.15

 Hubbard

0.15

 Nimbus

0.14

erin

0.14

_EDEFAULT

0.14

bean

0.14

asu

0.14

ÐµÑĢÐ¶

0.14

æ¸

0.13

Activations Density 0.004%

code snippets and programming constructs

No Comments

No Known Activations