INDEX

Explanations

references to various schemes or programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ship

-0.18

 McCabe

-0.15

most

-0.15

lene

-0.15

sure

-0.14

oma

-0.14

sson

-0.14

obus

-0.14

uce

-0.14

lee

-0.14

POSITIVE LOGITS

pered

0.21

antics

0.18

atically

0.18

urally

0.18

atics

0.16

forth

0.16

ely

0.16

cloth

0.16

yard

0.16

etat

0.15

Activations Density 0.010%

references to various schemes or programs

No Comments

No Known Activations