INDEX

Explanations

modal verbs indicating past actions or desires

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

los

-0.15

loi

-0.15

 shared

-0.15

ŀæĢ§

-0.15

 possibly

-0.14

ancell

-0.14

uggle

-0.14

isy

-0.13

 shar

-0.13

 Shared

-0.13

POSITIVE LOGITS

 often

0.21

often

0.19

 sometimes

0.18

sometimes

0.17

 occasionally

0.16

iec

0.15

ouz

0.15

 whenever

0.15

Often

0.15

 Often

0.15

Activations Density 0.050%

modal verbs indicating past actions or desires

No Comments

No Known Activations