INDEX

Explanations

words related to various types of services offered

oai_token-act-pair · gpt-3.5-turbo

references to various types of services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ross

-0.69

Filename

-0.66

uran

-0.65

lihood

-0.65

Ze

-0.64

NING

-0.62

izon

-0.61

 Kern

-0.61

rik

-0.60

iary

-0.59

POSITIVE LOGITS

 rendered

1.11

 provider

1.03

 providers

1.00

 Provider

0.99

cape

0.86

 Provided

0.86

essions

0.75

oteric

0.74

pring

0.74

ourcing

0.74

Activations Density 0.039%

words related to various types of services offered

references to various types of services

No Comments

No Known Activations