INDEX

Explanations

phrases or terms related to the word "row."

oai_token-act-pair · gpt-3.5-turbo

references to "Row" followed by a number, indicating a focus on rows in a dataset or structured information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ¤

-1.01

urated

-0.78

 blown

-0.71

ãĥ¼ãĥĨãĤ£

-0.70

UNE

-0.68

urate

-0.67

urations

-0.67

 Rouhani

-0.66

ural

-0.66

uration

-0.63

POSITIVE LOGITS

dy

1.45

dies

1.39

nd

0.99

ser

0.94

een

0.90

zee

0.89

hani

0.85

lands

0.85

land

0.82

orld

0.80

Activations Density 0.036%

phrases or terms related to the word "row."

references to "Row" followed by a number, indicating a focus on rows in a dataset or structured information

No Comments

No Known Activations