INDEX

Explanations

references to specific individuals and their opinions in a political context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Experiment

-0.17

 experimental

-0.16

Experiment

-0.15

 Experimental

-0.15

-scrollbar

-0.15

 experiment

-0.14

æıı

-0.13

experimental

-0.13

uento

-0.13

ori

-0.13

POSITIVE LOGITS

 meant

0.27

 included

0.19

 cherry

0.19

 INCLUDED

0.18

esh

0.18

 forgot

0.18

 word

0.17

included

0.17

 pref

0.17

 Included

0.17

Activations Density 0.300%

references to specific individuals and their opinions in a political context

No Comments

No Known Activations