INDEX

Explanations

instances of the word "who" and its related forms in the context of personal descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

abbo

-0.16

tti

-0.14

ternet

-0.14

ABEL

-0.14

igua

-0.14

ateria

-0.13

anske

-0.13

NÃį

-0.13

Responder

-0.13

irie

-0.13

POSITIVE LOGITS

 himself

0.18

 whom

0.16

jezd

0.15

 earlier

0.15

 Fach

0.15

imli

0.14

osh

0.14

0.13

 Earlier

0.13

 ìŀĦ

0.13

Activations Density 0.103%

instances of the word "who" and its related forms in the context of personal descriptions

No Comments

No Known Activations