INDEX

Explanations

names or titles within sentences, such as people or locations

oai_token-act-pair · gpt-3.5-turbo

specific names, titles, or terms related to locations and individuals within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

thumbnails

-0.71

-0.69

touch

-0.68

essa

-0.67

uses

-0.66

EMP

-0.65

iss

-0.64

 Sally

-0.63

ISO

-0.63

POSITIVE LOGITS

0.90

Bs

0.90

 Barkley

0.89

Bib

0.88

 Byrd

0.88

 Bung

0.87

Gob

0.85

 Bret

0.85

 Berg

0.84

Activations Density 0.253%

names or titles within sentences, such as people or locations

specific names, titles, or terms related to locations and individuals within a context

No Comments

No Known Activations