INDEX

Explanations

phrases that relate to legal claims or assertions

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

_Print

-0.16

.Resume

-0.16

remen

-0.15

npj

-0.15

mite

-0.14

.transport

-0.14

ije

-0.13

:animated

-0.13

arena

-0.13

POSITIVE LOGITS

 claim

0.16

 claimed

0.16

 claims

0.16

Virgin

0.15

Lum

0.15

baugh

0.15

Pow

0.15

izzo

0.15

/disc

0.14

ÅĪ

0.14

Activations Density 0.037%

phrases that relate to legal claims or assertions

No Comments

No Known Activations