INDEX

Explanations

phrases related to escalating or increasing

oai_token-act-pair · gpt-3.5-turbo

instances of the word "up"

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

cut

-0.66

====

-0.62

Anne

-0.61

ow

-0.58

alf

-0.57

INO

-0.57

 Abyss

-0.57

Eye

-0.57

Closure

-0.57

lean

-0.57

POSITIVE LOGITS

river

0.90

stairs

0.79

grading

0.78

dates

0.78

raised

0.77

hol

0.77

 shop

0.76

grades

0.76

votes

0.74

 nicely

0.71

Activations Density 0.127%

phrases related to escalating or increasing

instances of the word "up"

No Comments

No Known Activations