INDEX

Explanations

technical terms related to access tokens and authentication processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

³

-0.15

 Thur

-0.15

 Trilogy

-0.15

mites

-0.14

 Baba

-0.14

 Troy

-0.13

 dipl

-0.13

 Wikispecies

-0.13

éīĦ

-0.13

Kup

-0.13

POSITIVE LOGITS

 token

0.74

 Token

0.65

token

0.63

 tokens

0.61

Token

0.61

-token

0.59

_token

0.59

.token

0.58

(token

0.55

 TOKEN

0.55

Activations Density 0.034%

technical terms related to access tokens and authentication processes

No Comments

No Known Activations