INDEX

Explanations

phrases related to things staying the same or not being affected

oai_token-act-pair · gpt-3.5-turbo

terms related to stability or lack of change

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ç«

-0.68

alez

-0.67

 Typhoon

-0.67

aph

-0.65

RH

-0.62

 McKenna

-0.60

=-=-=-=-=-=-=-=-

-0.60

ingo

-0.60

Â¯Â¯

-0.60

eur

-0.59

POSITIVE LOGITS

 unchanged

1.24

 untouched

0.85

 unaffected

0.83

iated

0.80

ishment

0.74

theless

0.73

aneously

0.71

ãĤ´

0.70

iating

0.68

 interpol

0.67

Activations Density 0.006%

phrases related to things staying the same or not being affected

terms related to stability or lack of change

No Known Activations