INDEX

Explanations

conjunctions and phrases denoting connections or relationships between concepts

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

its

-0.15

atch

-0.14

_OLD

-0.13

_suspend

-0.13

lements

-0.13

 pozn

-0.13

.Factory

-0.13

ause

-0.13

sut

-0.13

-0.12

POSITIVE LOGITS

 extent

0.21

/or

0.20

 entirety

0.19

 nature

0.19

 spirit

0.16

ivery

0.16

ĵn

0.16

 duration

0.15

 confines

0.15

 tone

0.15

Activations Density 0.111%

conjunctions and phrases denoting connections or relationships between concepts

No Comments

No Known Activations