INDEX

Explanations

words related to actions or behavior that have an intentional or purposeful quality

oai_token-act-pair · gpt-3.5-turbo

words related to various forms of scientific and analytical attributes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olicited

-0.90

owship

-0.83

dust

-0.76

anca

-0.74

mbuds

-0.72

allows

-0.72

under

-0.72

IRD

-0.71

CAN

-0.70

aiman

-0.69

POSITIVE LOGITS

ative

1.13

atives

0.73

terday

0.72

 minded

0.70

 Warfare

0.69

ãĥ¼ãĥĨ

0.66

 ende

0.66

 Measures

0.66

 interrogation

0.66

ATIVE

0.66

Activations Density 0.019%

words related to actions or behavior that have an intentional or purposeful quality

words related to various forms of scientific and analytical attributes

No Comments

No Known Activations