INDEX

Explanations

instances of repetition or similarity in phrases or descriptions

oai_token-act-pair · gpt-3.5-turbo

words associated with correctness or appropriateness 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 @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ļéĨĴ

-0.81

antam

-0.73

ntil

-0.69

ADRA

-0.68

ibal

-0.68

quished

-0.67

ĳ

-0.64

âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦

-0.63

Introdu

-0.63

Minimum

-0.62

POSITIVE LOGITS

 money

0.75

 smells

0.70

 messenger

0.70

 manner

0.69

 livelihood

0.68

 timetable

0.68

 histories

0.68

 manners

0.68

 colors

0.67

 geography

0.66

Activations Density 0.717%

instances of repetition or similarity in phrases or descriptions

words associated with correctness or appropriateness in various contexts

No Comments

No Known Activations