INDEX

Explanations

references to significant concepts or quantities in academic or scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Ã¹ng

-0.19

íı°

-0.14

urre

-0.14

_VC

-0.14

èĥ

-0.14

íĴ

-0.14

 MessageType

-0.13

rente

-0.13

aments

-0.13

canf

-0.13

POSITIVE LOGITS

phan

0.17

íĭ´

0.15

UGE

0.15

 Lois

0.15

0.14

 extension

0.14

phy

0.14

 Linden

0.14

.defaults

0.14

 trembling

0.14

Activations Density 0.001%

references to significant concepts or quantities in academic or scientific contexts

No Comments

No Known Activations