INDEX

Explanations

terms and phrases related to categories, midpoints, and gradations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

olas

-0.15

ceil

-0.15

engin

-0.15

 Geile

-0.14

ÏģÎ¿Ïħ

-0.14

ahn

-0.14

imas

-0.14

_rgba

-0.13

ÑĭÑĪ

-0.13

ä¼´

-0.13

POSITIVE LOGITS

 intermediate

0.57

 middle

0.55

 Intermediate

0.47

middle

0.44

Intermediate

0.44

 intermediary

0.43

 between

0.40

 intermedi

0.39

ä¸Ń

0.39

 midd

0.39

Activations Density 0.267%

terms and phrases related to categories, midpoints, and gradations in various contexts

No Comments

No Known Activations