INDEX

Explanations

references to specific items or concepts indicated by the word "this."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

aji

-0.15

scoped

-0.15

elda

-0.15

_levels

-0.14

oret

-0.14

 shelf

-0.14

 doorstep

-0.14

omu

-0.14

Ð²Ð¾ÑĢ

-0.14

å¤ªéĥİ

-0.14

POSITIVE LOGITS

 regard

0.47

 regards

0.38

 instance

0.33

 context

0.33

 manner

0.31

 vein

0.31

 respect

0.31

 respects

0.30

 sense

0.29

 capacity

0.29

Activations Density 0.055%

references to specific items or concepts indicated by the word "this."

No Comments

No Known Activations