INDEX

Explanations

terms related to network technology and encryption protocols

oai_token-act-pair · gpt-3.5-turbo

technical specifications and performance metrics of technology products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

undo

-0.77

sylv

-0.73

pedia

-0.73

naire

-0.72

john

-0.70

jong

-0.67

borgh

-0.66

makers

-0.65

frey

-0.64

ulas

-0.64

POSITIVE LOGITS

ailability

0.73

 decoding

0.72

Enc

0.71

 compliant

0.70

WithNo

0.66

 looph

0.64

CBC

0.64

 unbeliev

0.63

ENC

0.62

 acknow

0.61

Activations Density 0.071%

terms related to network technology and encryption protocols

technical specifications and performance metrics of technology products

No Comments

No Known Activations