INDEX

Explanations

terms related to locations or establishments with the word "way" in them

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "way" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

uster

-0.90

icio

-0.76

ividual

-0.71

itatively

-0.71

anguage

-0.71

usters

-0.68

 tremend

-0.67

utical

-0.65

anches

-0.65

rador

-0.64

POSITIVE LOGITS

fare

1.24

finding

1.12

bill

1.09

ward

0.96

point

0.94

WARD

0.92

PASS

0.88

way

0.87

finder

0.82

frame

0.81

Activations Density 0.040%

terms related to locations or establishments with the word "way" in them

occurrences of the word "way" in various forms

No Comments

No Known Activations