INDEX

Explanations

instances of the word "accepted"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "accepted" in various contexts

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

iture

-0.80

iov

-0.67

abouts

-0.60

cffffcc

-0.59

Blu

-0.59

bing

-0.59

damage

-0.59

illon

-0.59

 bott

-0.59

Track

-0.58

POSITIVE LOGITS

 accepted

3.64

accept

2.07

 accepts

1.93

 accept

1.92

 Accept

1.89

Accept

1.81

 accepting

1.77

 rejected

1.71

 acceptance

1.68

 agreed

1.66

Activations Density 0.007%

instances of the word "accepted"

instances of the word "accepted" in various contexts

No Comments

No Known Activations