INDEX

Explanations

repeated variations of the letter "S" or similar patterns in text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

fort

-0.16

pur

-0.16

holm

-0.16

 NETWORK

-0.15

ogenic

-0.15

dah

-0.15

yre

-0.15

 bulld

-0.14

emark

-0.14

LOSS

-0.14

POSITIVE LOGITS

gart

0.17

ritel

0.15

mamak

0.15

fÃ¼h

0.14

ãĥĸãĥª

0.14

chers

0.14

Ã¹

0.14

regon

0.14

urette

0.14

æķ£

0.14

Activations Density 0.384%

repeated variations of the letter "S" or similar patterns in text

No Comments

No Known Activations