INDEX

Explanations

the word 'ken' with different numeric values

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "ken" in various contexts

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

accompan

-0.76

 pestic

-0.72

 cryst

-0.72

 trave

-0.71

 murd

-0.70

ENCY

-0.69

 Canaveral

-0.67

EB

-0.67

 skelet

-0.67

 exha

-0.67

POSITIVE LOGITS

ken

1.08

azi

1.07

ners

1.02

burgh

1.01

feld

0.96

wald

0.89

ning

0.89

hof

0.85

wich

0.85

heit

0.84

Activations Density 0.019%

the word 'ken' with different numeric values

occurrences of the term "ken" in various contexts

No Comments

No Known Activations