INDEX

Explanations

phrases indicating recognition or high regard for subjects in various contexts

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

insky

-0.17

nell

-0.14

 Myers

-0.14

mer

-0.14

/github

-0.13

ester

-0.13

abl

-0.13

Muk

-0.13

Î±Î¹

-0.13

 weakest

-0.13

POSITIVE LOGITS

igel

0.15

 amongst

0.15

 among

0.15

Incre

0.15

ATUS

0.14

 Andreas

0.14

 Among

0.14

ndata

0.14

ims

0.14

atk

0.14

Activations Density 0.203%

phrases indicating recognition or high regard for subjects in various contexts

No Comments

No Known Activations