INDEX

Explanations

terms related to specific identification or classification

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of "scientific" or "scientific" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

PRES

-0.77

CTV

-0.76

JP

-0.72

 EntityItem

-0.69

 Sach

-0.65

 largeDownload

-0.65

Ô

-0.64

iewicz

-0.64

©¶æ

-0.64

VK

-0.63

POSITIVE LOGITS

ific

1.29

atory

1.15

ature

1.09

ulty

1.07

ially

0.98

ual

0.93

ance

0.92

ulum

0.92

ates

0.92

atures

0.91

Activations Density 0.009%

terms related to specific identification or classification

words related to the concept of "scientific" or "scientific" in different contexts

No Comments

No Known Activations