INDEX

Explanations

negative descriptors associated with social and political commentary

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

reeze

-0.18

olib

-0.15

wan

-0.14

 bastante

-0.14

 quite

-0.14

 Quite

-0.14

pte

-0.14

idente

-0.14

atori

-0.14

Į

-0.14

POSITIVE LOGITS

 that

0.30

that

0.28

 ettÃ¤

0.22

 daÃŁ

0.22

 ÑĩÑĤÐ¾

0.22

	that

0.22

 ráº±ng

0.21

 Å¾e

0.21

 bahwa

0.20

 dass

0.20

Activations Density 0.118%

negative descriptors associated with social and political commentary

No Comments

No Known Activations