INDEX

Explanations

references to the concept of "about" or ideas discussing relationships and connections among various themes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

rava

-0.18

imo

-0.15

aylight

-0.15

odia

-0.15

ylon

-0.14

hal

-0.14

ez

-0.14

ursors

-0.14

chez

-0.14

",__

-0.14

POSITIVE LOGITS

 ways

0.28

how

0.21

all

0.20

 possible

0.20

 whom

0.18

who

0.18

 possibilities

0.17

 Ways

0.17

 implications

0.17

 repercussions

0.17

Activations Density 0.076%

references to the concept of "about" or ideas discussing relationships and connections among various themes

No Comments

No Known Activations