INDEX

Explanations

instances of the word "encounter" and its variations to identify discussions of interactions or meetings

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

uden

-0.18

ÙĪÙĨØ§

-0.17

igi

-0.16

undler

-0.15

readcrumb

-0.14

oco

-0.14

Ð¾ÑĩÐºÑĥ

-0.14

ansom

-0.14

ocaly

-0.14

ÐºÐ¾Ð½

-0.13

POSITIVE LOGITS

inge

0.19

 encounter

0.16

 encountered

0.16

icut

0.16

lush

0.15

inges

0.15

 Carlson

0.15

 encount

0.14

 Gros

0.14

 Ð¿ÑĢÐµÐ¿

0.14

Activations Density 0.031%

instances of the word "encounter" and its variations to identify discussions of interactions or meetings

No Comments

No Known Activations