INDEX

Explanations

locations and dates where someone was involved in an event

oai_token-act-pair · gpt-3.5-turbo

instances of the verb "was" 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 @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

-------------

-0.75

ettings

-0.73

lein

-0.72

feel

-0.70

 Extend

-0.68

Which

-0.67

rule

-0.66

cheat

-0.66

rates

-0.66

Â¢

-0.66

POSITIVE LOGITS

 originally

1.27

 flanked

1.16

 reportedly

1.14

 photographed

1.10

 scheduled

1.08

 spotted

1.07

 arrested

1.07

 hailed

1.07

 slated

1.06

 initially

1.05

Activations Density 0.191%

locations and dates where someone was involved in an event

instances of the verb "was" in various contexts

No Comments

No Known Activations