INDEX

Explanations

instances of the phrase "in a" followed by other contextual words

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

kinson

-0.14

Ple

-0.14

 Inform

-0.14

Ãªu

-0.14

 Investigation

-0.14

roz

-0.14

Ð¸ÑģÑĤÑĢÐ°

-0.13

 Bureau

-0.13

 entire

-0.13

POSITIVE LOGITS

 statement

0.24

 prepared

0.21

statement

0.20

 emailed

0.19

prepared

0.18

 email

0.18

	statement

0.17

 interview

0.15

 blog

0.15

 written

0.15

Activations Density 0.032%

instances of the phrase "in a" followed by other contextual words

No Comments

No Known Activations