INDEX

Explanations

themes related to comparisons and preferences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

hsi

-0.16

/Sub

-0.15

@Spring

-0.15

 addslashes

-0.14

soup

-0.14

IGHL

-0.14

sam

-0.14

Å¡ti

-0.13

sheets

-0.13

/Sh

-0.13

POSITIVE LOGITS

1.14

0.71

ÂłS

0.64

 Ø³

0.59

=S

0.59

_s

0.57

:S

0.56

.getS

0.54

 getS

0.53

.s

0.52

Activations Density 0.275%

themes related to comparisons and preferences

No Comments

No Known Activations