INDEX

Explanations

phrases related to learning and gaining knowledge

oai_token-act-pair · gpt-3.5-turbo

references to learning experiences and important lessons

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

uxe

-0.71

igham

-0.68

cohol

-0.67

lished

-0.66

ospace

-0.65

oples

-0.65

omo

-0.64

vertisement

-0.62

raviolet

-0.61

athi

-0.61

POSITIVE LOGITS

 lesson

0.98

 firsthand

0.91

 lessons

0.79

 skills

0.73

 Skills

0.73

>(

0.71

 anew

0.71

 tricks

0.71

İĭ

0.70

 basics

0.69

Activations Density 0.129%

phrases related to learning and gaining knowledge

references to learning experiences and important lessons

No Comments

No Known Activations