INDEX

Explanations

references to academic journal volume and issue numbers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

atural

-0.15

pte

-0.15

atten

-0.15

iÅŁleri

-0.14

hammer

-0.14

 Vital

-0.14

OLUMN

-0.14

$http

-0.13

atic

-0.13

 Merr

-0.13

POSITIVE LOGITS

/gcc

0.16

loat

0.15

ekim

0.15

riere

0.15

ãĥ¥

0.15

ç¾

0.14

ãĥ³ãĤº

0.13

occo

0.13

 Shapiro

0.13

rms

0.13

Activations Density 0.011%

references to academic journal volume and issue numbers

No Comments

No Known Activations