INDEX

Explanations

references to political figures and their associated actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

mb

-0.15

argon

-0.15

alker

-0.14

avig

-0.14

 instrument

-0.14

MB

-0.14

 hollow

-0.14

andin

-0.14

ationale

-0.14

otions

-0.14

POSITIVE LOGITS

ti

0.21

sis

0.20

tain

0.20

si

0.19

ture

0.19

tures

0.19

tu

0.18

bil

0.18

ni

0.17

tle

0.17

Activations Density 0.012%

references to political figures and their associated actions

No Comments

No Known Activations