INDEX

Explanations

specific details or information mentioned in sentences

oai_token-act-pair · gpt-3.5-turbo

questions and inquiries about unspecified information or details

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.82

Think

-0.73

imum

-0.71

½

-0.71

goo

-0.69

Imagine

-0.65

 mediocre

-0.65

hei

-0.64

 creeps

-0.64

little

-0.63

POSITIVE LOGITS

 confir

0.95

 specifics

0.89

nor

0.87

 affiliation

0.80

 involvement

0.73

 formally

0.71

 extent

0.71

 motive

0.69

 reimburse

0.69

 specific

0.67

Activations Density 0.384%

specific details or information mentioned in sentences

questions and inquiries about unspecified information or details

No Comments

No Known Activations