INDEX

Explanations

references to female characters and their emotions or actions

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

å¦»

-0.27

 wife

-0.26

 Wife

-0.22

/she

-0.21

wife

-0.21

 himself

-0.20

 ÙĨÙģØ³Ùĩ

-0.19

sing

-0.17

 seul

-0.16

ship

-0.16

POSITIVE LOGITS

/he

0.39

 herself

0.39

athed

0.36

pher

0.35

esh

0.33

pherd

0.30

ikh

0.30

ffield

0.27

athing

0.27

pard

0.26

Activations Density 0.147%

references to female characters and their emotions or actions

No Comments

No Known Activations