INDEX

Explanations

mentions of influence or being influenced by something or someone

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "influenced" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

amina

-0.80

ft

-0.78

VK

-0.73

clamation

-0.69

çĦ

-0.67

FT

-0.65

 Honour

-0.64

Letter

-0.62

Job

-0.62

chat

-0.61

POSITIVE LOGITS

 influenced

1.37

 influences

1.24

 influencing

1.07

 influence

1.03

 swayed

0.99

 sway

0.96

iasco

0.87

ierrez

0.85

 shaping

0.82

luence

0.81

Activations Density 0.010%

mentions of influence or being influenced by something or someone

occurrences of the word "influenced" and its variations

No Comments

No Known Activations