INDEX

Explanations

contrastive phrases indicating limitations or exceptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

.scalablytyped

-0.17

PLUS

-0.16

 zase

-0.15

/antlr

-0.15

shint

-0.15

ozÃŃ

-0.14

Å¡Å¥

-0.14

porter

-0.14

ColumnInfo

-0.14

ØºØ·

-0.14

POSITIVE LOGITS

 none

0.38

 nowhere

0.32

 None

0.28

none

0.28

 among

0.28

None

0.24

 NONE

0.24

among

0.24

 Among

0.24

Among

0.23

Activations Density 0.100%

contrastive phrases indicating limitations or exceptions

No Comments

No Known Activations