INDEX

Explanations

phrases related to historical events or outcomes

oai_token-act-pair · gpt-3.5-turbo

themes related to fate and historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

inka

-0.67

TAG

-0.67

cart

-0.65

ornia

-0.64

orget

-0.64

agascar

-0.61

assembly

-0.60

hitting

-0.60

oji

-0.60

itter

-0.60

POSITIVE LOGITS

 dictates

1.63

 dictated

1.25

 dictate

1.24

 forbids

1.17

 prevailed

1.17

 tells

1.15

 intervened

1.15

 wills

1.11

 permits

1.09

 justifies

1.07

Activations Density 0.213%

phrases related to historical events or outcomes

themes related to fate and historical context

No Comments

No Known Activations