INDEX

Explanations

significant nouns and terms related to prominent events or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

maz

-0.17

itos

-0.17

ildo

-0.15

 cancellation

-0.15

ohn

-0.15

ito

-0.15

-0.14

aura

-0.14

ResponseStatus

-0.14

amm

-0.14

POSITIVE LOGITS

RET

0.24

ret

0.23

Ret

0.23

ret

0.22

 retina

0.21

Ret

0.20

-ret

0.20

RET

0.19

_ret

0.17

ret

0.16

Activations Density 0.024%

significant nouns and terms related to prominent events or concepts

No Comments

No Known Activations