INDEX

Explanations

references to interpersonal relationships and social interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

alone

-0.17

 indeed

-0.16

 Alone

-0.16

iors

-0.15

icari

-0.15

inde

-0.15

 Indeed

-0.15

oran

-0.15

Indeed

-0.14

aille

-0.14

POSITIVE LOGITS

 again

0.35

again

0.29

 Again

0.26

 thereby

0.26

Again

0.25

 based

0.23

AGAIN

0.22

_again

0.21

ased

0.21

 AGAIN

0.20

Activations Density 0.015%

references to interpersonal relationships and social interactions

No Comments

No Known Activations