INDEX

Explanations

excerpts related to problem-solving or resolution

oai_token-act-pair · gpt-3.5-turbo

instances of the word "resolved"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ammy

-0.90

icipated

-0.86

ooks

-0.81

aquin

-0.72

appa

-0.71

older

-0.70

DI

-0.70

nesota

-0.69

inen

-0.69

amy

-0.67

POSITIVE LOGITS

 resolves

1.01

 resolving

1.00

 resolved

0.92

 resolve

0.90

actionDate

0.80

olution

0.76

 resolution

0.76

lucent

0.74

istant

0.72

ĵĺ

0.70

Activations Density 0.009%

excerpts related to problem-solving or resolution

instances of the word "resolved"

No Comments

No Known Activations