INDEX

Explanations

occurrences of the word "had" and other forms of possession or past experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

rape

-0.15

innie

-0.14

åŁĭ

-0.14

omat

-0.14

pir

-0.13

 Shame

-0.13

box

-0.13

orne

-0.13

 protest

-0.13

.manage

-0.13

POSITIVE LOGITS

 opportunity

0.16

 chance

0.15

 success

0.15

mdb

0.15

 surgery

0.15

 succes

0.15

 Banner

0.14

Ã©v

0.14

 difficulty

0.14

opi

0.14

Activations Density 0.071%

occurrences of the word "had" and other forms of possession or past experiences

No Comments

No Known Activations