INDEX

Explanations

occurrences of the word "spot" along with related words like "spotlight" or "spotchecking."

oai_token-act-pair · gpt-3.5-turbo

instances of the term "spot" 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 @ 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

issance

-0.76

 Strait

-0.67

 stride

-0.62

racuse

-0.60

 pend

-0.60

 adolesc

-0.59

yss

-0.59

wake

-0.59

 consortium

-0.58

 arcs

-0.57

POSITIVE LOGITS

lights

1.64

ty

1.26

ter

1.24

light

1.13

ters

1.09

lighting

1.09

tery

1.04

ting

1.01

pots

0.98

less

0.94

Activations Density 0.036%

occurrences of the word "spot" along with related words like "spotlight" or "spotchecking."

instances of the term "spot" in various contexts

No Comments

No Known Activations