INDEX

Explanations

instances of processes or events that have been ongoing or reflective in nature

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

usra

-0.17

ÙĤÛĮ

-0.16

akah

-0.15

ichick

-0.15

yen

-0.14

adaki

-0.14

kke

-0.14

ownik

-0.14

onian

-0.14

ãĥªãĤ¢

-0.13

POSITIVE LOGITS

 long

0.73

 years

0.57

long

0.54

 ages

0.48

 LONG

0.47

.long

0.45

 months

0.44

 YEARS

0.44

-long

0.43

 lÃ¢u

0.43

Activations Density 0.278%

instances of processes or events that have been ongoing or reflective in nature

No Comments

No Known Activations