INDEX

Explanations

the word "As" in various contexts, indicating a focus on comparative phrases or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

oes

-0.17

 evidenced

-0.17

tas

-0.17

oe

-0.16

activated

-0.16

WithURL

-0.16

ivec

-0.15

actly

-0.15

eat

-0.15

activate

-0.15

POSITIVE LOGITS

ylum

0.21

raf

0.18

gard

0.18

coli

0.18

untos

0.18

 soon

0.17

utral

0.17

far

0.17

enciÃ³n

0.17

mode

0.16

Activations Density 0.026%

the word "As" in various contexts, indicating a focus on comparative phrases or conditions

No Comments

No Known Activations