INDEX

Explanations

references to investigations or criminal activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

alla

-0.16

 eget

-0.15

 Fist

-0.15

optgroup

-0.15

urvey

-0.14

.backup

-0.14

allas

-0.14

 biá»ĥu

-0.14

opia

-0.14

cision

-0.14

POSITIVE LOGITS

DNA

0.28

 leads

0.23

DNA

0.21

 Leads

0.20

 reward

0.20

 Rewards

0.20

 solved

0.19

 tips

0.19

 Reward

0.19

dna

0.19

Activations Density 0.094%

references to investigations or criminal activities

No Comments

No Known Activations