INDEX

Explanations

terms related to research findings or outcomes

oai_token-act-pair · gpt-3.5-turbo

mentions of "results" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Passage

-0.72

resent

-0.67

 vows

-0.64

 vigil

-0.63

ModLoader

-0.60

 slog

-0.59

 actionGroup

-0.58

yne

-0.58

nt

-0.58

 lords

-0.58

POSITIVE LOGITS

results

0.89

iments

0.88

iveness

0.88

 results

0.86

 Results

0.77

ample

0.75

ivity

0.74

 thereof

0.74

Results

0.73

amples

0.73

Activations Density 0.025%

terms related to research findings or outcomes

mentions of "results" in various contexts

No Comments

No Known Activations