INDEX

Explanations

references to risk or danger, particularly in contexts that imply vulnerability or potential loss

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ÎºÏģÎ±

-0.16

andex

-0.15

anton

-0.15

ask

-0.15

elic

-0.15

ack

-0.15

blk

-0.14

aston

-0.14

acimiento

-0.14

undle

-0.14

POSITIVE LOGITS

atron

0.15

otron

0.15

daki

0.15

 Kirk

0.14

ê¸Ī

0.14

rouw

0.14

owaÄĩ

0.14

å¼ı

0.14

arium

0.14

seys

0.14

Activations Density 0.001%

references to risk or danger, particularly in contexts that imply vulnerability or potential loss

No Comments

No Known Activations