INDEX

Explanations

terms related to crossing, whether it be borders or paths in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

iko

-0.15

iments

-0.15

ernals

-0.15

.Interop

-0.15

cer

-0.15

uk

-0.14

ide

-0.14

ested

-0.14

etc

-0.14

cn

-0.14

POSITIVE LOGITS

roads

0.24

ses

0.23

-cross

0.19

-sectional

0.18

ylv

0.18

cross

0.17

inion

0.17

bow

0.16

cut

0.16

(es

0.15

Activations Density 0.024%

terms related to crossing, whether it be borders or paths in various contexts

No Comments

No Known Activations