INDEX

Explanations

specific references or instances of something within a broader context

oai_token-act-pair · gpt-3.5-turbo

references to "that" in various contexts, often inquiring about specifics or highlighting notable elements

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

ographies

-0.87

kamp

-0.85

izont

-0.84

engers

-0.84

ourses

-0.84

atures

-0.83

Methods

-0.83

eeds

-0.83

å§«

-0.81

emis

-0.80

POSITIVE LOGITS

 pesky

1.58

 fateful

1.15

 damned

1.12

 dreaded

1.12

 bastard

1.09

 elusive

1.07

 annoying

1.04

 darn

1.02

 sucker

1.01

 goofy

1.01

Activations Density 0.193%

specific references or instances of something within a broader context

references to "that" in various contexts, often inquiring about specifics or highlighting notable elements

No Comments

No Known Activations