INDEX

Explanations

phrases related to connections or associations

oai_token-act-pair · gpt-3.5-turbo

references to ties in various contexts, including emotional connections and competitive situations

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

sburg

-0.78

afa

-0.72

artney

-0.72

angel

-0.72

issance

-0.69

iation

-0.68

ouston

-0.66

agher

-0.66

azard

-0.64

iverpool

-0.64

POSITIVE LOGITS

tie

1.14

 knot

1.07

breaker

1.03

tie

1.00

 tied

0.93

pipe

0.92

breakers

0.87

 tying

0.87

 dyed

0.87

 knots

0.85

Activations Density 0.027%

phrases related to connections or associations

references to ties in various contexts, including emotional connections and competitive situations

No Comments

No Known Activations