INDEX

Explanations

variations of the word "alternate."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

yw

-0.19

æ°£

-0.17

æ°Ķ

-0.17

 Stevenson

-0.17

opr

-0.16

ipo

-0.16

ziej

-0.16

rada

-0.15

ewan

-0.15

yon

-0.15

POSITIVE LOGITS

 Kemal

0.16

chine

0.16

 distrust

0.15

 repmat

0.15

Äįet

0.15

sec

0.14

WN

0.14

oru

0.14

isan

0.14

Activations Density 0.018%

variations of the word "alternate."

No Comments

No Known Activations