INDEX

Explanations

words related to generalizing, standardizing, or applying a specific concept universally

oai_token-act-pair · gpt-3.5-turbo

words related to generalization and categorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Greenwich

-0.68

 Quinn

-0.67

=-=-=-=-

-0.65

ntil

-0.65

 bras

-0.65

ãĥ¼ãĥ«

-0.65

tes

-0.64

 Derby

-0.64

perty

-0.63

-0.62

POSITIVE LOGITS

Application

0.90

yip

0.89

 Correction

0.79

apply

0.79

otomy

0.77

opathic

0.74

application

0.73

ortment

0.70

iates

0.70

ensical

0.70

Activations Density 0.066%

words related to generalizing, standardizing, or applying a specific concept universally

words related to generalization and categorization

No Comments

No Known Activations