INDEX

Explanations

unique character and identity representations 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 @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

GRAM

-0.14

æİĪ

-0.14

bands

-0.13

ught

-0.13

 liberals

-0.13

argent

-0.13

/Public

-0.12

Ð¾ÑĢÑĸÐ²

-0.12

 fault

-0.12

lexport

-0.12

POSITIVE LOGITS

 epis

0.37

 episode

0.32

Ep

0.31

episode

0.30

 Episode

0.30

ep

0.30

 Ã©p

0.29

 episodes

0.29

Ep

0.26

Episode

0.26

Activations Density 0.007%

unique character and identity representations in text

No Comments

No Known Activations