INDEX

Explanations

words related to processing and altering materials or substances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

asted

-0.16

agedList

-0.15

ic

-0.15

edin

-0.14

ised

-0.14

iq

-0.14

herited

-0.14

bound

-0.13

 kontakt

-0.13

Mus

-0.13

POSITIVE LOGITS

ness

0.32

/generated

0.20

NESS

0.19

anken

0.17

icts

0.17

ijkstra

0.16

noch

0.15

ERENCE

0.15

amel

0.14

ecimal

0.14

Activations Density 0.741%

words related to processing and altering materials or substances

No Comments

No Known Activations