INDEX

Explanations

words associated with classifying or categorizing things

oai_token-act-pair · gpt-3.5-turbo

terms related to classification or categorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Archdemon

-0.71

 destro

-0.64

 shine

-0.63

 outweigh

-0.62

addin

-0.61

 pist

-0.61

 brisk

-0.61

 reperto

-0.60

 Sharif

-0.58

 Baal

-0.58

POSITIVE LOGITS

inctions

0.81

ategories

0.81

ategory

0.76

 guiActiveUnfocused

0.74

lishes

0.69

 "$:/

0.69

ulhu

0.69

icated

0.68

verages

0.68

ombat

0.67

Activations Density 0.365%

words associated with classifying or categorizing things

terms related to classification or categorization

No Comments

No Known Activations