INDEX

Explanations

phrases related to significant events or entities

oai_token-act-pair · gpt-3.5-turbo

the term "great" in various contexts

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

ople

-0.96

cling

-0.73

ilus

-0.73

endif

-0.71

gging

-0.70

©¶æ

-0.68

PD

-0.67

utical

-0.67

former

-0.66

NG

-0.66

POSITIVE LOGITS

 strides

0.91

sword

0.88

 apes

0.79

 boon

0.76

itud

0.74

 importance

0.70

 pains

0.70

 proportions

0.70

 Dane

0.70

 offence

0.69

Activations Density 0.034%

phrases related to significant events or entities

the term "great" in various contexts

No Comments

No Known Activations