INDEX

Explanations

phrases related to imitation, learning, and instruction

oai_token-act-pair · gpt-3.5-turbo

conjunctions and their association with various actions or states

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

ledged

-0.77

scribe

-0.75

arton

-0.71

pedia

-0.68

aceae

-0.66

rued

-0.66

igun

-0.66

lated

-0.65

cised

-0.65

ize

-0.65

POSITIVE LOGITS

 letting

1.72

 putting

1.68

 making

1.62

 delivering

1.61

 creating

1.60

 adapting

1.58

 gaining

1.58

 distributing

1.58

 getting

1.57

 discovering

1.57

Activations Density 0.352%

phrases related to imitation, learning, and instruction

conjunctions and their association with various actions or states

No Comments

No Known Activations