INDEX

Explanations

terms related to roller skating and roller coasters

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

aland

-0.15

pop

-0.15

 Sovere

-0.14

aka

-0.14

able

-0.14

akah

-0.14

vak

-0.14

aket

-0.14

 crown

-0.14

POSITIVE LOGITS

peater

0.15

endas

0.15

urg

0.15

IVED

0.14

incer

0.14

ÏģÎ¹Î¬

0.14

ãĥ¼ãĥĦ

0.14

chedulers

0.14

addtogroup

0.14

ëĮ

0.14

Activations Density 0.004%

terms related to roller skating and roller coasters

No Known Activations