INDEX

Explanations

references to numbered sections or lists in textual documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

sik

-0.16

ãģŁãĤĬ

-0.16

sburg

-0.15

hazi

-0.15

ulen

-0.15

Ð¸ÐºÐ¸

-0.15

OfType

-0.14

NSK

-0.14

UIL

-0.14

POSITIVE LOGITS

Wed

0.16

inois

0.15

IX

0.15

ouv

0.15

Wed

0.15

itin

0.14

ray

0.14

ly

0.14

brick

0.14

Activations Density 0.057%

references to numbered sections or lists in textual documents

No Comments

No Known Activations