INDEX

Explanations

words related to entertainment or content categorization

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

orer

-0.17

jsp

-0.17

endar

-0.17

 Ð³ÑĢÐ¾

-0.15

ORE

-0.14

OPY

-0.14

ãĥ«ãĥķ

-0.14

-Identifier

-0.14

icao

-0.14

 cedar

-0.14

POSITIVE LOGITS

kuk

0.16

cap

0.15

 Elliot

0.15

rix

0.14

_singleton

0.14

loff

0.14

RIX

0.14

unsch

0.13

 Carm

0.13

CLUDING

0.13

Activations Density 0.000%

words related to entertainment or content categorization

No Comments

No Known Activations