INDEX

Explanations

names of places and cultural references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

LETTE

-0.16

rana

-0.15

Â·Â·

-0.14

riba

-0.14

ilha

-0.14

cade

-0.13

iras

-0.13

orf

-0.13

/qt

-0.13

etten

-0.13

POSITIVE LOGITS

ìļķ

0.15

Ìģ

0.14

urance

0.14

 Toll

0.13

IMUM

0.13

 toll

0.13

-utils

0.13

ews

0.13

argout

0.13

Activations Density 1.022%

names of places and cultural references

No Comments

No Known Activations