INDEX

Explanations

terms indicating quantity or a sense of totality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

itler

-0.17

Ã¼ss

-0.15

à¥Ĥà¤·

-0.15

sg

-0.15

anova

-0.14

sw

-0.14

 Baron

-0.14

sr

-0.14

ITTER

-0.14

Kv

-0.14

POSITIVE LOGITS

ness

0.20

ãĥ³ãĤ¯

0.16

plies

0.15

/not

0.15

foreground

0.14

ogy

0.14

ood

0.14

ongan

0.14

oretical

0.13

Activations Density 0.385%

terms indicating quantity or a sense of totality

No Comments

No Known Activations