INDEX

Explanations

words associated with encyclopedic content or detailed informational formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

-0.16

avage

-0.16

 háºŃu

-0.14

scenes

-0.14

marsh

-0.14

sten

-0.14

ilities

-0.14

ehen

-0.14

rees

-0.13

 Trace

-0.13

POSITIVE LOGITS

lopedia

0.29

 Britann

0.28

edia

0.22

pedia

0.22

Brit

0.21

clo

0.21

opa

0.20

yclopedia

0.20

lical

0.19

 Britt

0.19

Activations Density 0.007%

words associated with encyclopedic content or detailed informational formats

No Comments

No Known Activations