INDEX

Explanations

adjectives or phrases describing attributes or characteristics of things

oai_token-act-pair · gpt-3.5-turbo

concepts related to strengths, weaknesses, and moral considerations within different 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

swick

-0.82

Ĥİ

-0.78

thouse

-0.68

UTE

-0.64

¥

-0.62

åī

-0.61

ulp

-0.60

 batches

-0.60

regor

-0.60

 Revival

-0.58

POSITIVE LOGITS

 comparable

0.88

 attached

0.84

 whatsoever

0.84

 lined

0.82

 implanted

0.79

 built

0.78

 similar

0.77

 ranging

0.76

 spanning

0.75

 knack

0.74

Activations Density 0.309%

adjectives or phrases describing attributes or characteristics of things

concepts related to strengths, weaknesses, and moral considerations within different contexts

No Comments

No Known Activations