INDEX

Explanations

words related to political figures and government positions

oai_token-act-pair · gpt-3.5-turbo

the presence of the term "Ex" which appears to indicate a specific context or subject matter within the document

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

 GOODMAN

-0.69

 Learns

-0.67

enegger

-0.67

 beginnings

-0.66

die

-0.64

jriwal

-0.64

 heights

-0.64

 haze

-0.63

 Sawyer

-0.61

WARE

-0.61

POSITIVE LOGITS

haust

1.27

cerpt

1.15

ternally

1.15

isting

1.12

changes

1.12

terior

1.12

uber

1.10

cessive

1.09

posing

1.05

pert

1.05

Activations Density 0.019%

words related to political figures and government positions

the presence of the term "Ex" which appears to indicate a specific context or subject matter within the document

No Comments

No Known Activations