INDEX

Explanations

mentions of the location "Tempe" in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Tempe" and its derivatives

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atform

-0.77

ãģį

-0.77

 Barkley

-0.72

Magikarp

-0.71

 Farage

-0.68

ASC

-0.66

 Scotia

-0.65

 Newfoundland

-0.64

 Ferdinand

-0.64

 Barcl

-0.63

POSITIVE LOGITS

poral

1.43

porary

1.21

ples

1.19

plates

1.17

pered

1.14

pering

1.12

perate

1.09

ension

1.04

tem

0.97

pora

0.97

Activations Density 0.012%

mentions of the location "Tempe" in various contexts

occurrences of the word "Tempe" and its derivatives

No Comments

No Known Activations