INDEX

Explanations

instances where something is presented or officially shown

oai_token-act-pair · gpt-3.5-turbo

instances of the word "present" and its variations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ï¸ı

-0.77

cffff

-0.74

terness

-0.71

ionage

-0.67

imb

-0.67

kie

-0.65

leen

-0.64

bel

-0.64

icka

-0.64

 tweeting

-0.64

POSITIVE LOGITS

 slides

0.75

 evidence

0.75

 Present

0.73

 presentations

0.73

Ĥª

0.72

Solution

0.71

 presents

0.71

 Exhibit

0.70

evidence

0.67

eering

0.67

Activations Density 0.059%

instances where something is presented or officially shown

instances of the word "present" and its variations

No Comments

No Known Activations