INDEX

Explanations

phrases related to unexpected discoveries or challenges

oai_token-act-pair · gpt-3.5-turbo

instances of the word "stumble" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

cius

-0.73

folios

-0.66

cell

-0.63

Ã±a

-0.61

atu

-0.60

rera

-0.59

=#

-0.58

nesota

-0.58

verts

-0.57

cers

-0.57

POSITIVE LOGITS

 stumble

0.84

weed

0.83

 upon

0.82

Upon

0.79

worthiness

0.76

upon

0.75

 stumbling

0.74

 blindly

0.73

INST

0.70

bum

0.69

Activations Density 0.059%

phrases related to unexpected discoveries or challenges

instances of the word "stumble" in various forms

No Comments

No Known Activations