INDEX

Explanations

references to disgust and annoyance in experiences or descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 Collapse

-0.17

 collapse

-0.15

 failures

-0.14

 collapses

-0.14

 collapsed

-0.14

-collapse

-0.13

collapsed

-0.13

å¼±

-0.13

ÑģÑĥÑĤ

-0.13

elta

-0.13

POSITIVE LOGITS

 annoyance

0.50

 annoying

0.49

 annoy

0.49

 irritating

0.47

 bothers

0.45

 nuisance

0.43

 irritation

0.42

 nuis

0.41

 anno

0.41

 irrit

0.41

Activations Density 0.376%

references to disgust and annoyance in experiences or descriptions

No Comments

No Known Activations