INDEX

Explanations

locations or geographic features

oai_token-act-pair · gpt-3.5-turbo

geographical locations and related features

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

ptr

-0.82

options

-0.81

kw

-0.75

é¾

-0.74

»Ĵ

-0.71

cells

-0.70

capacity

-0.70

hillary

-0.69

tackle

-0.67

fam

-0.66

POSITIVE LOGITS

 Telescope

0.89

 Observatory

0.87

 Institute

0.86

 Directorate

0.82

 Scrolls

0.82

 Disorder

0.80

 Mountains

0.80

 Depot

0.79

 Syndrome

0.78

 Initiative

0.77

Activations Density 0.106%

locations or geographic features

geographical locations and related features

No Comments

No Known Activations