INDEX

Explanations

phrases that discuss performance ratings or comparisons in film and game reviews

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

VOKE

-0.15

cestor

-0.15

SpaceItem

-0.15

 Albums

-0.14

 porad

-0.14

igh

-0.14

ceph

-0.14

adoras

-0.14

odont

-0.14

antage

-0.14

POSITIVE LOGITS

 cast

0.21

 photography

0.16

 dial

0.16

 chemistry

0.16

 music

0.15

 supporting

0.15

 FixedUpdate

0.15

 female

0.15

 technical

0.15

 secondary

0.15

Activations Density 0.201%

phrases that discuss performance ratings or comparisons in film and game reviews

No Comments

No Known Activations