INDEX

Explanations

occurrences of the word "hat" in various forms and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

quat

-0.18

scape

-0.16

eron

-0.16

tank

-0.16

etto

-0.16

bett

-0.15

sdale

-0.15

eration

-0.15

gress

-0.15

usercontent

-0.15

POSITIVE LOGITS

 trick

0.25

rack

0.23

red

0.23

ting

0.22

field

0.22

 Trick

0.21

 tricks

0.21

reds

0.20

RED

0.20

rick

0.20

Activations Density 0.009%

occurrences of the word "hat" in various forms and contexts

No Comments

No Known Activations