INDEX

Explanations

words related to physical locations, specifically stations

oai_token-act-pair · gpt-3.5-turbo

references to various types of stations, particularly in transportation contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

aphael

-0.73

 Hague

-0.72

URES

-0.72

 IMAGES

-0.68

 Resolution

-0.66

 Kinnikuman

-0.64

 Coalition

-0.64

FLAG

-0.63

 Pain

-0.63

Gam

-0.62

POSITIVE LOGITS

 stations

1.47

 station

1.19

 wagon

1.04

wagon

0.95

igmatic

0.81

mith

0.80

hare

0.79

ngth

0.78

 lapt

0.76

 terminals

0.75

Activations Density 0.012%

words related to physical locations, specifically stations

references to various types of stations, particularly in transportation contexts

No Comments

No Known Activations