INDEX

Explanations

specific terms related to medical, legal, and compliance matters

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

 Stranger

-0.17

andez

-0.15

ninger

-0.14

uste

-0.14

gere

-0.14

endid

-0.14

 newcomer

-0.14

 Garner

-0.14

emer

-0.14

ospital

-0.14

POSITIVE LOGITS

 users

0.25

0.20

 participants

0.20

users

0.18

 folk

0.18

çĶ¨æĪ·

0.17

 patients

0.17

_users

0.16

0.15

 holders

0.15

Activations Density 0.360%

specific terms related to medical, legal, and compliance matters

No Comments

No Known Activations