INDEX

Explanations

phrases related to topics like technology, health, and lifestyle

oai_token-act-pair · gpt-3.5-turbo

mentions of additional items or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

xtap

-0.77

era

-0.74

pload

-0.67

Gra

-0.67

Enlarge

-0.66

 Rite

-0.65

gee

-0.64

 Mellon

-0.63

steen

-0.62

-0.61

POSITIVE LOGITS

 importantly

1.50

 than

0.96

 consequ

0.89

 sophisticated

0.88

 refined

0.86

 challeng

0.82

 elaborate

0.81

 advanced

0.81

 stringent

0.81

 efficient

0.80

Activations Density 0.059%

phrases related to topics like technology, health, and lifestyle

mentions of additional items or concepts

No Comments

No Known Activations