INDEX

Explanations

HTML markup tags

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

&r

-0.16

éļ

-0.14

opes

-0.14

ILI

-0.14

coe

-0.14

ataka

-0.13

/command

-0.13

 diam

-0.13

ift

-0.13

allee

-0.13

POSITIVE LOGITS

-placeholder

0.15

enburg

0.14

atha

0.14

EXTERN

0.14

engu

0.14

hots

0.14

Importer

0.14

kyt

0.14

edor

0.14

Activations Density 0.016%

HTML markup tags

No Comments

No Known Activations