INDEX

Explanations

references to crime and legal cases, particularly those involving murder

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

aed

-0.16

erior

-0.15

erm

-0.15

.lu

-0.14

erv

-0.14

ernal

-0.13

edian

-0.13

nad

-0.13

ortal

-0.13

anti

-0.13

POSITIVE LOGITS

@brief

0.15

istrovstvÃŃ

0.14

okit

0.14

lub

0.14

haft

0.14

ÑĩÐ°ÑĤ

0.14

OVÃģ

0.14

VERTISE

0.13

¯u

0.13

 Masks

0.13

Activations Density 0.102%

references to crime and legal cases, particularly those involving murder

No Comments

No Known Activations

references to crime and legal cases, particularly those involving murder

No Comments

No Known Activations