INDEX

Explanations

names of companies and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

pron

-0.15

ruba

-0.14

hape

-0.14

Ã½t

-0.14

jid

-0.14

ouri

-0.13

emez

-0.13

ÑĢÐ¾Ð½

-0.13

ÅŁt

-0.13

hari

-0.13

POSITIVE LOGITS

avra

0.16

illis

0.15

avaÅŁ

0.13

argar

0.13

uras

0.13

 maduras

0.13

aton

0.12

 Jensen

0.12

(poly

0.12

 itself

0.12

Activations Density 0.036%

names of companies and organizations

No Comments

No Known Activations