INDEX

Explanations

references to deputy-level positions in organizations or government

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

pNet

-0.16

otta

-0.16

oub

-0.15

 ÑĩÐ¸Ð½

-0.15

ãĤĮãģ©

-0.15

issors

-0.15

åĴ¨

-0.14

 actionTypes

-0.14

roje

-0.14

inesis

-0.14

POSITIVE LOGITS

ship

0.18

 mayor

0.18

 editor

0.18

dom

0.17

-editor

0.17

editor

0.17

 marshal

0.17

ships

0.16

 Editor

0.16

 sheriff

0.15

Activations Density 0.011%

references to deputy-level positions in organizations or government

No Comments

No Known Activations