INDEX

Explanations

observations or instances of observing

oai_token-act-pair · gpt-3.5-turbo

references to the act of observing or conducting observations

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

interstitial

-0.77

ikuman

-0.76

cit

-0.76

gur

-0.76

recy

-0.73

lins

-0.69

boss

-0.69

ifice

-0.67

umble

-0.67

zzy

-0.66

POSITIVE LOGITS

 observation

0.96

 Observ

0.87

 observing

0.79

 patterns

0.72

 observable

0.72

 observed

0.71

 observe

0.70

 observations

0.69

 observer

0.69

lihood

0.67

Activations Density 0.042%

observations or instances of observing

references to the act of observing or conducting observations

No Comments

No Known Activations