INDEX

Explanations

words related to camping or outdoor activities, with a focus on tents

oai_token-act-pair · gpt-3.5-turbo

references to tents or related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ÃĽ

-0.68

EVA

-0.67

 symmetry

-0.66

bye

-0.64

 congr

-0.62

nesota

-0.61

Words

-0.61

obbies

-0.60

Second

-0.60

 Cuomo

-0.59

POSITIVE LOGITS

atively

1.34

acle

1.29

pole

1.21

acles

1.02

 encamp

0.97

 tent

0.96

 tents

0.90

acular

0.88

 pole

0.88

 poles

0.86

Activations Density 0.015%

words related to camping or outdoor activities, with a focus on tents

references to tents or related concepts

No Comments

No Known Activations