INDEX

Explanations

repeated patterns of the word "my" and its variations, indicating a focus on personal possessive expressions

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

akk

-0.17

Linear

-0.15

 Shield

-0.14

mouseleave

-0.14

shield

-0.14

fullscreen

-0.14

Fit

-0.14

.Utc

-0.14

à¤¾à¤¸à¤¨

-0.14

POSITIVE LOGITS

zen

0.15

£

0.15

loff

0.15

 åľ°

0.14

ovy

0.14

leston

0.14

illy

0.13

Bomb

0.13

ÃĹ↵↵

0.13

ocos

0.13

Activations Density 0.006%

repeated patterns of the word "my" and its variations, indicating a focus on personal possessive expressions

No Comments

No Known Activations