INDEX

Explanations

names related to a specific location

oai_token-act-pair · gpt-3.5-turbo

references to specific names or terms related to places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 horm

-0.65

grounds

-0.63

 STATS

-0.61

 crop

-0.60

 steroid

-0.60

 arenas

-0.59

 hoops

-0.59

 bubbles

-0.58

 Disneyland

-0.58

bc

-0.58

POSITIVE LOGITS

itu

0.73

lycer

0.71

hai

0.71

Wan

0.69

 Toast

0.68

hower

0.67

ahi

0.66

asury

0.66

zee

0.64

Fei

0.63

Activations Density 0.142%

names related to a specific location

references to specific names or terms related to places

No Comments

No Known Activations