INDEX

Explanations

occurrences of the word "from" related to various contexts of removal or withdrawal

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

cheid

-0.15

lÃŃÄį

-0.14

èµ·æĿ¥

-0.14

elop

-0.13

ctl

-0.13

 Regel

-0.13

kan

-0.13

rary

-0.13

uite

-0.13

@Spring

-0.13

POSITIVE LOGITS

 sight

0.30

 altogether

0.25

 circulation

0.22

 khá»ıi

0.21

 view

0.21

 involvement

0.21

 being

0.21

 existence

0.19

 consideration

0.19

 Sight

0.19

Activations Density 0.130%

occurrences of the word "from" related to various contexts of removal or withdrawal

No Comments

No Known Activations