INDEX

Explanations

prepositions 'against a' followed by a noun

oai_token-act-pair · gpt-3.5-turbo

articles and adjectives that describe objects or entities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ACTION

-0.86

liest

-0.79

Events

-0.78

Hours

-0.77

 associations

-0.74

OTOS

-0.74

oms

-0.71

Reviewer

-0.71

 Cases

-0.71

effective

-0.71

POSITIVE LOGITS

 cliff

1.13

 boulder

1.06

 pile

1.04

 tree

1.03

 fence

1.02

 doorway

1.00

 wall

0.99

 pillar

0.98

 ledge

0.97

 hill

0.97

Activations Density 0.165%

prepositions 'against a' followed by a noun

articles and adjectives that describe objects or entities

No Comments

No Known Activations