INDEX

Explanations

occurrences of the word "voted" in various contexts and forms

oai_token-act-pair · gpt-3.5-turbo

instances of the word 'voted'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 intermedi

-0.63

 attribution

-0.62

duration

-0.60

 caution

-0.59

 technique

-0.59

 Peak

-0.59

 overhead

-0.57

kin

-0.56

 recovery

-0.56

 tracing

-0.55

POSITIVE LOGITS

 voted

3.91

 vote

2.35

 votes

2.23

 voting

2.04

vote

1.99

 Voting

1.93

Vote

1.91

 Vote

1.90

 Votes

1.87

 ballots

1.61

Activations Density 0.011%

occurrences of the word "voted" in various contexts and forms

instances of the word 'voted'

No Comments

No Known Activations