INDEX

Explanations

verbs or phrases describing actions and occurrences

oai_token-act-pair · gpt-3.5-turbo

phrases indicating possession or the existence of issues

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

 Uses

-0.69

 Discuss

-0.68

 Choose

-0.67

Review

-0.63

 consult

-0.62

 hears

-0.62

 Pair

-0.61

 chooses

-0.60

 Refer

-0.59

 Comprehensive

-0.59

POSITIVE LOGITS

 resulted

1.35

 prompted

1.19

 been

1.19

 caused

1.15

 coincided

1.15

led

1.14

 culminated

1.11

 ensured

1.11

 eroded

1.11

 sparked

1.07

Activations Density 0.116%

verbs or phrases describing actions and occurrences

phrases indicating possession or the existence of issues

No Comments

No Known Activations