INDEX

Explanations

expressions of personal identity and individual experience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ÑģÑĤÐ¸

-0.14

ervas

-0.14

apiro

-0.14

ÑĸÐ´Ð½Ð¾

-0.14

.annotate

-0.13

arios

-0.13

ãĥ£

-0.13

 sortOrder

-0.12

assi

-0.12

Æ¡

-0.12

POSITIVE LOGITS

 talking

0.75

 referring

0.73

 refer

0.69

 refers

0.60

 Talking

0.56

 talk

0.55

Talking

0.54

 reference

0.53

refer

0.49

 speaking

0.48

Activations Density 0.152%

expressions of personal identity and individual experience

No Comments

No Known Activations