INDEX

Explanations

instances where someone or something is being "put" in a specific situation or context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "put" in various contexts

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

externalActionCode

-0.87

 Austral

-0.67

BLE

-0.63

Flight

-0.62

cale

-0.61

 Hear

-0.60

reader

-0.60

riott

-0.60

 UNCLASSIFIED

-0.58

 Grounds

-0.58

POSITIVE LOGITS

tering

0.96

tin

0.94

downs

0.92

ative

0.86

ter

0.83

ongh

0.81

ty

0.80

down

0.80

tered

0.79

away

0.79

Activations Density 0.032%

instances where someone or something is being "put" in a specific situation or context

instances of the word "put" in various contexts

No Comments

No Known Activations