INDEX

Explanations

phrases and words that indicate requests for information and strong emotional stakes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

Ã³b

-0.16

ikan

-0.16

aldo

-0.16

-0.14

assen

-0.14

é£

-0.14

ady

-0.14

sed

-0.14

alan

-0.14

san

-0.13

POSITIVE LOGITS

anzeigen

0.16

yz

0.15

 Band

0.15

 Hamm

0.15

aukee

0.14

aux

0.14

Band

0.14

pmat

0.14

xes

0.13

ÛĮÙĨÚ©

0.13

Activations Density 0.000%

phrases and words that indicate requests for information and strong emotional stakes

No Comments

No Known Activations