INDEX

Explanations

terms related to suspenseful situations or narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

ahlen

-0.17

/inet

-0.16

statt

-0.15

bers

-0.14

æµ·éģĵ

-0.14

à¸Ħà¸§à¸²à¸¡à¸Ħ

-0.14

antino

-0.14

á»¯a

-0.14

ollah

-0.13

essler

-0.13

POSITIVE LOGITS

argent

0.15

rio

0.15

onder

0.14

ãĤ«ãĥ¼

0.14

curities

0.14

 intact

0.14

RYPTO

0.14

umas

0.13

YLON

0.13

 porte

0.13

Activations Density 0.004%

terms related to suspenseful situations or narratives

No Comments

No Known Activations