INDEX

Explanations

terms associated with legal proceedings and prosecution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

_hooks

-0.16

 Legisl

-0.16

ÃŁe

-0.15

ritch

-0.14

swe

-0.14

 hurd

-0.14

 Arbit

-0.14

Å¯l

-0.14

íį¼

-0.14

 imageSize

-0.14

POSITIVE LOGITS

 office

0.18

 charging

0.18

unch

0.17

 cases

0.16

 Office

0.16

charging

0.15

DAG

0.15

 Charging

0.15

AG

0.15

agas

0.15

Activations Density 0.066%

terms associated with legal proceedings and prosecution

No Comments

No Known Activations