INDEX

Explanations

references to the term "Pan"

oai_token-act-pair · gpt-3.5-turbo

mentions of "Pan American" or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

itability

-0.64

 minded

-0.59

enance

-0.58

BILITIES

-0.58

 groom

-0.58

ï¸ı

-0.57

itably

-0.57

 foul

-0.57

roy

-0.56

 proced

-0.55

POSITIVE LOGITS

orama

1.25

asonic

1.13

acea

1.12

aman

1.12

tera

1.10

zers

1.07

ahi

1.07

etta

1.06

amera

1.05

zar

1.04

Activations Density 0.032%

references to the term "Pan"

mentions of "Pan American" or related terms

No Comments

No Known Activations