INDEX

Explanations

quantifiers and superlative adjectives indicating excessive amounts or degrees

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

adius

-0.18

åĪ°åºķ

-0.15

é¤Ĭ

-0.15

eler

-0.15

ntag

-0.15

oad

-0.14

ANJI

-0.14

cia

-0.14

iders

-0.14

 DISCLAIM

-0.14

POSITIVE LOGITS

TOO

0.24

Too

0.22

too

0.22

too

0.22

Too

0.20

-too

0.19

å¤ª

0.18

-than

0.15

/to

0.15

ilio

0.15

Activations Density 0.120%

quantifiers and superlative adjectives indicating excessive amounts or degrees

No Comments

No Known Activations