INDEX

Explanations

phrases that emphasize and address the reader directly, creating a sense of engagement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

led

-0.16

ersed

-0.16

iang

-0.15

sy

-0.15

ter

-0.15

xp

-0.15

et

-0.15

_userdata

-0.14

akis

-0.14

POSITIVE LOGITS

ãĥ«ãĥĪ

0.16

ÐŁÐļ

0.15

 Hayden

0.14

rub

0.14

 veloc

0.13

ateg

0.13

ubbo

0.13

letics

0.13

upos

0.13

Forgot

0.13

Activations Density 0.046%

phrases that emphasize and address the reader directly, creating a sense of engagement

No Comments

No Known Activations