INDEX

Explanations

references to organizational affiliations or associations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ever

-0.19

ey

-0.19

tas

-0.19

eat

-0.18

swire

-0.16

activate

-0.16

 evidenced

-0.16

jected

-0.16

spiel

-0.16

ighton

-0.15

POSITIVE LOGITS

coli

0.19

pects

0.17

utral

0.17

LOAT

0.17

ylum

0.16

sembl

0.16

untos

0.15

far

0.15

raf

0.15

ari

0.15

Activations Density 0.046%

references to organizational affiliations or associations

No Comments

No Known Activations