INDEX

Explanations

words related to implications or potential consequences of a situation

oai_token-act-pair · gpt-3.5-turbo

references to consequences or effects related to various topics

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

vez

-0.74

oned

-0.73

ced

-0.70

cop

-0.70

walk

-0.68

ves

-0.68

bows

-0.67

bones

-0.67

yards

-0.67

few

-0.66

POSITIVE LOGITS

 implications

1.07

 ramifications

0.89

romeda

0.82

 Impl

0.80

 fallout

0.79

notation

0.78

uality

0.77

notations

0.77

urities

0.76

ogene

0.76

Activations Density 0.024%

words related to implications or potential consequences of a situation

references to consequences or effects related to various topics

No Comments

No Known Activations