INDEX

Explanations

phrases indicating reasoning or causation

oai_token-act-pair · gpt-3.5-turbo

situations involving motivations and justifications for actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

roups

-0.93

enture

-0.85

ahime

-0.81

united

-0.75

olutions

-0.74

iversity

-0.73

sbm

-0.72

comings

-0.71

itect

-0.71

Our

-0.70

POSITIVE LOGITS

he

1.40

she

1.19

his

1.13

him

1.03

 hers

0.98

his

0.90

it

0.88

her

0.88

 himself

0.88

 accidentally

0.88

Activations Density 0.514%

phrases indicating reasoning or causation

situations involving motivations and justifications for actions

No Comments

No Known Activations