INDEX

Explanations

references to evidence or proof in a text

oai_token-act-pair · gpt-3.5-turbo

references to evidence or proof in discussions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

occ

-0.81

zinski

-0.69

roup

-0.67

":["

-0.65

Osw

-0.63

atche

-0.62

ade

-0.61

rik

-0.61

ades

-0.60

imar

-0.60

POSITIVE LOGITS

 proof

4.06

 Proof

3.03

Proof

2.83

 proofs

2.73

proof

2.69

 evidence

1.77

evidence

1.64

 proving

1.60

 evid

1.41

 verification

1.40

Activations Density 0.005%

references to evidence or proof in a text

references to evidence or proof in discussions

No Comments

No Known Activations