INDEX

Explanations

phrases related to legal or administrative actions and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ilater

-0.16

oui

-0.14

ÃĹ↵↵

-0.14

-0.13

Äįas

-0.13

alar

-0.13

ales

-0.13

cul

-0.13

POSITIVE LOGITS

antically

0.14

ãĤ£

0.14

uman

0.14

ichel

0.14

Ã©rÃ©

0.14

ettle

0.14

ãĥĥãĥģ

0.14

prav

0.14

AVING

0.14

ugins

0.14

Activations Density 0.021%

phrases related to legal or administrative actions and processes

No Comments

No Known Activations

phrases related to legal or administrative actions and processes

No Comments

No Known Activations