INDEX

Explanations

references to additional people or entities involved in a topic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

nea

-0.15

anda

-0.14

acier

-0.14

amer

-0.14

odega

-0.14

ÄĽla

-0.14

á»ĳi

-0.14

ank

-0.13

Ĭ

-0.13

 flexGrow

-0.13

POSITIVE LOGITS

idd

0.16

kees

0.15

tah

0.15

hol

0.15

Hol

0.14

Skip

0.14

cie

0.14

Bid

0.14

 anytime

0.14

yb

0.13

Activations Density 0.020%

references to additional people or entities involved in a topic

No Comments

No Known Activations