INDEX

Explanations

mentions or references to the word "design"

oai_token-act-pair · gpt-3.5-turbo

mentions of the term "design" and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Lauder

-0.81

ega

-0.77

ICAN

-0.77

 Cele

-0.77

rican

-0.67

ieri

-0.67

nikov

-0.66

ngth

-0.64

selves

-0.63

 Sheen

-0.63

POSITIVE LOGITS

ating

1.10

ated

1.10

ators

1.09

ations

1.07

ator

1.06

 aesthetic

0.96

ates

0.93

 flaws

0.92

yout

0.92

 flaw

0.90

Activations Density 0.066%

mentions or references to the word "design"

mentions of the term "design" and its various contexts

No Comments

No Known Activations