INDEX

Explanations

phrases related to a person named "Walker"

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the name "Walker."

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

ritic

-0.92

 Seym

-0.83

undai

-0.82

eon

-0.81

 Lumpur

-0.78

ION

-0.77

Ð½

-0.77

sembly

-0.77

iated

-0.75

unal

-0.73

POSITIVE LOGITS

inson

0.80

Walker

0.78

 Walker

0.76

esque

0.75

ball

0.74

stown

0.72

erness

0.69

tip

0.68

king

0.68

chairs

0.67

Activations Density 0.019%

phrases related to a person named "Walker"

repeated mentions of the name "Walker."

No Comments

No Known Activations