INDEX

Explanations

references to viral trends and memes on social media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

por

-0.16

esome

-0.15

odcast

-0.14

ween

-0.14

ognito

-0.14

 silver

-0.14

/scripts

-0.13

arest

-0.13

 hopefully

-0.13

lake

-0.13

POSITIVE LOGITS

Spread

0.16

NCY

0.16

GMEM

0.15

 viral

0.15

edla

0.15

brig

0.15

efon

0.14

å¾ª

0.14

$MESS

0.14

igon

0.14

Activations Density 0.017%

references to viral trends and memes on social media

No Comments

No Known Activations